Universidade do MinhoInstituto de Investigação em Biomateriais, Biodegradáveis e Biomiméticos

Mariana Rodrigues de Carvalho

setembro de 2019

Tissue Engineered In Vitro Models on a Chip for Cancer Research

Mar

iana

Rod

rigue

s de

Car

valh

o T

issu

e E

ng

ine

ere

d In V

itro

Mo

de

ls o

n a

Ch

ip

for

Ca

nce

r R

ese

arc

h

UM

inho

|201

9

Mariana Rodrigues de Carvalho

setembro de 2019

Tissue Engineered In Vitro Models on a Chip for Cancer Research

Trabalho efetuado sob a orientação do Professor Doutor Rui Luís Gonçalves dos Reis e doDoutor Joaquim Miguel Antunes Correia de Oliveira

Tese de Doutoramento em Engenharia de Tecidos, Medicina Regenerativa e Células Estaminais

Universidade do MinhoInstituto de Investigação em Biomateriais, Biodegradáveis e Biomiméticos

ii

DIREITOS DE AUTOR E CONDIÇÕES DE UTILIZAÇÃO DO TRABALHO POR TERCEIROS

Este é um trabalho académico que pode ser utilizado por terceiros desde que respeitadas as

regras e boas práticas internacionalmente aceites, no que concerne aos direitos de autor e direitos

conexos.

Assim, o presente trabalho pode ser utilizado nos termos previstos na licença abaixo indicada.

Caso o utilizador necessite de permissão para poder fazer um uso do trabalho em condições não

previstas no licenciamento indicado, deverá contactar o autor, através do RepositóriUM da Universidade

do Minho.

Licença concedida aos utilizadores deste trabalho

Atribuição CC BY

https://creativecommons.org/licenses/by/4.0/

iii

ACKNOWLEDGMENTS

The preparation of this thesis would not be possible without the important contribution of several

people, to which I am and always will be deeply thankful.

I would like to start by expressing my gratitude to my formal PhD supervisor, Prof. Rui Reis. He

gave me the opportunity to join his research group as a researcher, and later as a PhD student, always

supporting my work and progress. I would like to specially acknowledge his support and trust on me, for

his huge assistance in the PhD scholarship application, and also for allowing me to work with such good

co-supervisors. A special word of appreciation and gratitude goes to my co-supervisor, Dr. Joaquim

Miguel Oliveira. Since the beginning I felt he accepted me and trusted me, as a person and as a

researcher. Sometimes even acting as friend, always advising resilience, patience and hard work. A

great person to share a beer after a conference or to discuss the best path of research, who

understands the problems of life and of research. He trusted me with responsibilities and challenges,

with goals that I didn’t even know I could achieve. A big thank you to Dr. Roman Truckenmüller, Dr.

Stephan Giselbrecht and Dr. Pamela Habibovic for giving me the opportunity to work at MERLN

Institute, at Maastricht University. This world class institute provided me everything I needed to perform

an excellent work. Especially friends such as David Barata, Liliana Teixeira and David Boaventura, who

made my life richer, both professionally and personally. I am very grateful to my colleagues and friends

in 3B’s research group. We shared the good, the bad and the worst moments together, frustrations and

achievements. They know who they are! A huge and special thank you to Raquel Maia, who I met

during the “death valley” of my PhD, who basically thought me everything I know on science, thought

me how to think and work rigorously. Thank you for always being bossy but always with a smile and a

word of encouragement! To David Caballero, who joined me in a late phase of my PhD but gave me

ideas and inputs that helped immensely. It was a great pleasure to work and learn from you. For all of

these and much more, thank you David!

My truly and deeply-felt thanks to my parents, for being my safe harbor, my rock during the good

and bad moments, for giving me much more than I could ever want or need, for making me the person

I am today. I Love you so much! A big thank you to my twinnie sister, the best friend anyone could ever

ask for, a part of me. Also, I can’t forget to thank João, my new brother and faithful friend, always ready

to help!

Finally, I acknowledge my PhD scholarship NORTE-08-5369-FSE-000044, funded by Programa

Operacional Regional do Norte, Fundo Social Europeu, Norte2020 for the financial support.

iv

STATEMENT OF INTEGRITY

I hereby declare having conducted this academic work with integrity. I confirm that I have not used

plagiarism or any form of undue use of information or falsification of results along the process leading

to its elaboration.

I further declare that I have fully acknowledged the Code of Ethical Conduct of the University of

Minho.

v

ABSTRACT

Tissue Engineered In Vitro Models On A Chip For Cancer Research

By 2030, the global burden is expected to grow to 21.7 million new cancer cases and 13 million

cancer deaths simply due to the growth and aging of the population. Among all types of cancer,

colorectal cancer is a major cause of morbidity and mortality worldwide, and accounts for over 9 % of all

cancer incidence. It is the third most common cancer worldwide and affects men and women equally.

In order to win the battle against cancer, further advances are in great need to unveil identification

of cancer-causing agents in in vitro and in vivo animal models, as well as for the development of

personalized therapies, drug screening, and to provide insightful knowledge on the mechanisms of

tumor growth and metastasis. Microfluidic devices, together with tissue engineering strategies and

nanotechnology have emerged as a powerful platform to tackle the previously mentioned hurdles. These

themes are the focus of Section 1, in Chapters I, II and III.

In this thesis we focus on the application of CMCht/PAMAM dendrimer nanoparticles.as the

synthesis, uptake efficiency/internalization and cytotoxic effect of fluorescent-labeled CMCht/PAMAM

dendrimer nanoparticles was investigated using different cancer cell lines in both traditional culture

flasks and under physiological flow inside a microfluidic platform (Chapter V). Other than nanoparticles,

the use of different biomaterials to modulate 3D microenvironment of tumors is of the utmost

importance. Therefore, the use of HRP-crosslinked SF hydrogels with spatial tunable properties was

proposed in a colorectal cancer extravasation 3D model on a commercial chip Vena4™ (Chapter VI). To

increase the complexity of the cancer models, a microfluidic chip was designed and fabricated for the

incorporation of tumor and vascular parts, allowing for gradients of gemcitabine released from

CMCht/PAMAM nanoparticles to be tested (Chapter VII). Chemical modification in the CMCht/PAMAM

dendrimer nanoparticles in order to target colorectal cancer was also achieved in Chapter VIII. In

Chapter IX, a new proof of concept consisting of a microfluidic silk platform, flexible and implantable,

was developed in house.

The results and platforms developed in this thesis are discussed in Chapter X and represent a

strong advance in the field of lab-on-chip research and will be a useful tool for drug discovery and study

of migration/metastasis phenomena, allowing for a versatile choice of tissues, biomaterials and

biological assays.

Keywords: Biomaterials, Cancer models, Colorectal cancer, Microfluidics, Nanoparticles.

vi

RESUMO

Modelos In Vitro Em Chips Baseados Na Engenharia De Tecidos Para Investigação Na Área Do

Cancro

Até 2030, o número global de novos casos de cancro deverá crescer para 21,7 milhões, com

cerca de 13 milhões de mortes, simplesmente devido ao crescimento e envelhecimento da população.

Entre todos os tipos de cancro, o cancro colorretal é uma das principais causas de morbidade e

mortalidade em todo o mundo, representando mais de 9 % entre todos os tipos de cancro. Com o

objectivo de ganhar a guerra contra o cancro, novos avanços são rapidamente precisos, a fim de

identificar, em modelos animais in vitro e in vivo os agentes causadores de cancro, bem como para o

desenvolvimento de terapias personalizadas, teste de novos fármacos e aumentar o conhecimento

actual sobre os mecanismos de crescimento tumoral e metástases. Dispositivos de microfluídica

juntamente com estratégias de engenharia de tecidos e nanotecnologia surgiram como uma poderosa

plataforma para lidar com os obstáculos mencionados anteriormente. Esses temas são o foco da

Secção 1, nos Capítulos I, II e III. Nesta tese damos enfâse à aplicação de nanopartículas

CMCht/PAMAM. A síntese, eficiência, internalização e efeito citotóxico das mesmas foram investigados

usando diferentes linhas de células tumorais em frascos de cultura tradicionais e dentro de uma

plataforma microfluídica (Capítulo V). Além das nanopartículas, o uso de diferentes biomateriais para

modular o microambiente 3D de tumores é de extrema importância. Portanto, o uso de hidrogéis SF

reticulados com HRP com propriedades mecânicas ajustáveis foi proposto num modelo 3D de

extravasamento de cancro colorretal num chip comercial VenaT4™ (Capítulo VI). Para aumentar a

complexidade dos modelos de cancro em chips, uma plataforma microfluídica foi projetada e fabricada

para a incorporação de ambas as partes tumorais e vasculares, permitindo testar gradientes de

gemcitabina libertados das nanopartículas CMCht/PAMAM (Capítulo VII). Modificações químicas nas

nanopartículas para atingir especificamente células cancerígenas também foram alcançadas no

Capítulo VIII. No Capítulo IX, uma nova prova de conceito consistindo numa plataforma microfluídica de

seda, totalmente flexível e implantável, foi desenvolvida. Os resultados e plataformas desenvolvidos

nesta tese representam um forte avanço no campo da investigação do cancro, e serão uma ferramenta

útil para a descoberta de drogas e estudo de metastases, permitindo uma escolha versátil de tecidos,

hidrogéis e ensaios biológicos.

Palavras-chave: Biomateriais, Cancro colorrectal, Microfluídica, Modelos de cancro,

Nanopartículas.

vii

TABLE OF CONTENTS

ACKNOWLEDGMENTS ................................................................................................................... III

STATEMENT OF INTEGRITY ........................................................................................................... IV

ABSTRACT .....................................................................................................................................V

RESUMO ....................................................................................................................................... VI

TABLE OF CONTENTS .................................................................................................................. VII

LIST OF ABBREVIATIONS ...............................................................................................................XI

LIST OF EQUATIONS ................................................................................................................... XIX

LIST OF FIGURES .........................................................................................................................XX

LIST OF SUPPLEMENTARY FIGURES ......................................................................................... XXXI

LIST OF TABLES ....................................................................................................................... XXXII

SHORT CURRICULUM VITAE.................................................................................................... XXXIII

LIST OF PUBLICATIONS .......................................................................................................... XXXIV

INTRODUCTION TO THE THESIS FORMAT ............................................................................... XXXIX

SECTION 1 .................................................................................................................................... 1

GENERAL INTRODUCTION ............................................................................................................. 1

CHAPTER I - EVALUATING BIOMATERIALS AND MICROFLUIDICS BASED 3D TUMOR MODELS1 . 3

I-1. Abstract ......................................................................................................................... 3

I-2. The importance of 3D in vitro tissue models for advanced cancer research ..................... 4

I-3. Classical 3D TE models for cancer research ................................................................... 5

I-4. Chip based 3D TE models in cancer research .............................................................. 10

I-5. Concluding remarks and future perspectives ................................................................ 16

I-6. References .................................................................................................................. 17

CHAPTER II - BIOMATERIALS AND MICROFLUIDICS FOR DRUG DISCOVERY AND

DEVELOPMENT1 ...................................................................................................................... 24

Abstract .................................................................................................................................... 24

II-1. Introduction ............................................................................................................. 25

II-2. Biomaterials applied in microfluidic systems ............................................................ 27

II-3. Microfluidics in drug discovery ................................................................................. 30

viii

II-4. Organ-on-a-chip and drug discovery .......................................................................... 32

II-5. Conclusion and future trends ................................................................................... 39

II-6. References .............................................................................................................. 40

CHAPTER III - DENDRIMER NANOPARTICLES IN CRC APPLICATIONS ....................................... 45

Abstract .................................................................................................................................... 45

III-1. Introduction ............................................................................................................. 46

III-2. Dendrimer nanoparticles.......................................................................................... 48

III-3. Dendrimer applications in cancer research ............................................................... 52

III-4. Dendimer Nanoparticles in Colorectal Cancer........................................................... 54

III-5. Conclusions ............................................................................................................. 57

III-6. References .............................................................................................................. 58

SECTION 2 .................................................................................................................................. 62

MATERIALS AND METHODS ........................................................................................................ 62

CHAPTER IV - MATERIALS AND METHODS ................................................................................ 64

Overview ................................................................................................................................... 64

IV-1. Materials ................................................................................................................. 65

IV-2. Reagents ................................................................................................................. 77

IV-3. Methodologies for processing of biomaterials ........................................................... 77

IV-4. Dendrimer nanoparticle production .......................................................................... 77

IV-5. Hydrogels preparation ............................................................................................. 83

IV-6. Microfabrication ....................................................................................................... 86

IV-7. Physicochemical Characterization Techniques .......................................................... 92

IV-8. In vitro biological testing ........................................................................................ 102

IV-9. Statistical Analysis ................................................................................................. 124

IV-10. References ............................................................................................................ 125

SECTION 3 ................................................................................................................................ 133

EXPERIMENTAL SECTION .......................................................................................................... 133

CHAPTER V - A SEMIAUTOMATED MICROFLUIDIC PLATFORM FOR REAL TIME

INVESTIGATION OF NANOPARTICLES’ CELLULAR UPTAKE AND CANCER CELL’S TRACKING1 ........ 135

Abstract .................................................................................................................................. 135

ix

V-1. Graphical Abstract ................................................................................................. 136

V-2. Introduction ........................................................................................................... 136

V-3. Materials and Methods .......................................................................................... 139

V-4. Results .................................................................................................................. 147

V-5. Discussion ............................................................................................................. 154

V-6. Conclusions and future perspectives ...................................................................... 159

V-7. Acknowledgments .................................................................................................. 160

V-8. References ............................................................................................................ 160

CHAPTER VI - TUNING ENZYMATICALLY-CROSSLINKED SILK FIBROIN HYDROGEL

PROPERTIES FOR THE DEVELOPMENT OF A COLORECTAL CANCER EXTRAVASATION 3D MODEL ON

A CHIP1 .................................................................................................................. 164

Abstract .................................................................................................................................. 164

VI-1. Graphical Abstract ................................................................................................. 165

VI-2. Introduction ........................................................................................................... 165

VI-3. Materials and methods .......................................................................................... 168

VI-4. Results .................................................................................................................. 173

VI-5. Discussion ............................................................................................................. 182

VI-6. Conclusions ........................................................................................................... 187

VI-7. Acknowledgments .................................................................................................. 188

VI-8. References ............................................................................................................ 188

CHAPTER VII - COLORECTAL TUMOR-ON-A-CHIP SYSTEM: A 3D TOOL FOR PRECISION ONCO-

NANOMEDICINE1 ....................................................................................................................... 192

Abstract .................................................................................................................................. 192

VII-1. Graphical Abstract ................................................................................................. 193

VII-2. Introduction ........................................................................................................... 193

VII-3. Materials and Methods .......................................................................................... 195

VII-4. Results .................................................................................................................. 203

VII-5. Discussion ............................................................................................................. 211

VII-6. Conclusions ........................................................................................................... 217

VII-7. Acknowledgments .................................................................................................. 217

VII-8. References ............................................................................................................ 218

x

CHAPTER VIII - PEPTIDE-MODIFIED DENDRIMER NANOPARTICLES FOR TARGETED THERAPY OF

COLORECTAL CANCER1 .................................................................................................................. 222

Abstract .................................................................................................................................. 222

VIII-1. Graphical Abstract ................................................................................................. 223

VIII-2. Introduction ........................................................................................................... 223

VIII-3. Materials and Methods .......................................................................................... 225

VIII-4. Results and Discussion .......................................................................................... 231

VIII-5. Conclusions ........................................................................................................... 244

VIII-6. Acknowledgments .................................................................................................. 244

VIII-7. References ............................................................................................................ 245

VIII-8. Supplementary Figure ............................................................................................ 246

CHAPTER IX - PATENT - ENZYMATICALLY CROSSLINKED SILK FIBROIN HYDROGEL

MICROFLUIDIC PLATFORM, METHODS OF PRODUCTION AND USES THEREOF ............................. 250

Abstract .................................................................................................................................. 250

IX-1. Technical field ....................................................................................................... 251

IX-2. Background ........................................................................................................... 251

IX-3. General description ................................................................................................ 253

IX-4. Brief description of the drawings ............................................................................ 255

IX-5. Detailed description ............................................................................................... 256

IX-6. Claims................................................................................................................... 261

IX-7. List of Figures ........................................................................................................ 265

IX-8. References ............................................................................................................ 269

SECTION 4 ................................................................................................................................ 271

CONCLUSIONS AND FUTURE PERSPECTIVES ........................................................................... 271

CHAPTER X - CONCLUSIONS AND FUTURE PERSPECTIVES ................................................... 273

X-1. General conclusions .............................................................................................. 273

X-2. Future research directions ..................................................................................... 276

xi

LIST OF ABBREVIATIONS

α – Alpha

AB – Alamar blue

Abs – Absorbance

ADSCs – Adipose-derived stem cells

AFM – Atomic force microscopy

AgNPs – Silver nanoparticles

ANOVA – Analysis of variance

APPACDM – The Portuguese Association of Parents and Friends of Mentally Disabled Citizens

ASCs – Adipose tissue derived stromal cells

ASGPRs – Asialoglycoprotein receptors

ATR – FTIR – Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy

Au – Gold

AuNPs – Gold nanoparticles

AuNRs – Gold nanorods

B

β – Beta

β-sheet – Beta-sheet

BMSCs – Bone marrow derived mesenchymal stromal cells

BSA – Bovine serum albumin

C

CA – Chitosan-alginate scaffolds

Ca2+ – Calcium ions

Calcein-AM – Calcein-Acetoxymethy

CAM – Chick chorioallantoic membrane

CDDP – cis-diamine platinum

CDF – curcumin derivative

CHT – Chitosan

Cl- – Chloride ion

cm – centimeter

cm-1 – Reciprocal wavelength centimeters

xii

CMCht – CarboxymethylChitosan

CMCht/PAMAM – Carboxymethylchitosan/poly(amidoamine) dendrimers

CAN – Compartmentalized neuronal array

CNS – Central nervous system

CNT – Carbon nanotubes

CO2 – Carbon dioxide

Cp – Heat capacity

CRC – Colorectal cancer

CTCs – Circulating tumor cells

D

°C – Degree Celsius

d – Days

Da – Dalton

DA – Degree of acetylation

DAPI – 4',6-diamidino-2-phenylindole

DGAV – Direcção Geral de Alimentação e Veterinária

DLS – Dynamic light scattering

DMEM – Dulbecco’s modified Eagle’s medium

DMSO – Dimethylsulfoxide

DNA – Deoxyribonucleic acid

DOX – Doxorubicin

DS – degree of substitution

DSC – Differential scanning calorimetry

2D – 2-Dimensions

3D – 3-Dimensions

E

E – Young modulus

e.g. – For example, from latin exempli gratia

E’ – Elastic modulus

ECs – Endothelial cells

ECM – Extracellular matrix

EDA – Ethylenediamine

xiii

EDC – 1-Ethyl-(3-dimethylaminopropyl) carbodiimide hydrochloride

EDTA – Ethylenediaminetetraacetic acid

EHS – Engelbreth-Holm-Swarm

ELISA – Enzyme-linked immunosorbent Assay

EPR – Enhanced permeation and retention effect

Eq – Equation

EO – Ethylene oxide gas

ESB – European society for biomaterials

eSF – Enzymatically-crosslinked silk fibroin

et al. – And others

Ex/Em – Exitation/Emission

F

FBS – Fetal bovine serum

FDA – Food and drug administration

Fig. – Figure

FITC – Fluorescein isothiocyanate

FRA – Folate receptor-α

FTIR – Fourier transform infrared spectroscopy

G

g – grams

G1 – Generation 1

G2 – Generation 2

G3 -- Generation 2

G′ – Storage modulus

G′′ – Loss modulus

GAGs – Glycosaminoglycans

GEM – Gemcitabine

GFs – Growth factors

GG – Gellan gum

H

H – Hours

xiv

HCC – Hepatocellular cellular carcinoma

HeLa – Cervical cancer cells

H&E – Hematoxylin–eosin

H2O2 – Hydrogen peroxide

HA – Hyaluronic acid

HA-GG – High acyl gellan gum

HBMSCs – Human bone marrow derived mesenchymal stromal cells

HCl – Hydrorochloric acid

HCO32- – Bicarbonate ion

HCOMECS – Human colonic microvascular endothelial cells

HCT-119 – Colorectal cancer cell line

HFIP - Hexafluoroisopropanol

H1 NMR – Proton nuclear magnetic resonance

HRP – Horseradish peroxidase

Hz – Hertz

I

i.e. – “In other words”, from latin id est

IL – Interleukin

iPSCs – Induced pluripotent stem cells

K

K+ – Potassium ion

KBr – Potassium bromide

kDa – KiloDalton

kg – Kilogram

kN – kiloNewton

kPa – KiloPascal

L

L – Liters

L929 – Murine fibroblast cell line

LA-GG – Low acyl gellan gum

LiBr – Lithium bromide

xv

LNCaP – Lymph node carcinoma of the prostate

LT – Long-term

LVR – Linear viscoelastic region

M

m – meter

M – Molar

mA – Milliampere

MA-GG – Methacrylated gellan gum

MBP – Myelin basic protein

MCP-1 – Monocyte chemoattractant protein 1

MeCbl – Metil cobalamin

MEM – Minimum Essential Medium

MES - 2-(N-morpholino)ethanesulfonic acid

mg – milligram

Mg2+ – Magnesium ion

MenSCs – Human menstrual MSCs

Micro-CT – Micro computed tomography

Min – minute

miRNA – MicroRNAs

mL – Milliliter

mm – Millimeter

mM – Millimolar

Mf – Final mass

MMw – Medium molecular weight

Mw – Wet mass

MMP – Metalloproteinase

MNPs – Magnetic nanoparticles

Mol – Mole

MPa – megaPascal

MRI – Magnetic resonance image

MSCs – Mesenchymal derived stromal cells

MSCs – Mesenchymal stem cells

mw – Wet weight

xvi

MWCO – Molecular weight cut off

µm – micrometer

µ-PIV – Microparticle image velocimetry

N

n – Number of samples

N/A – Not applicable, Not available

Na+ – Sodium ion

NaCl – Sodium chloride

NaOH – Sodium hydroxide

NCMC - N-carboxymethyl chitosan

NGF – Nerve growth factor

NIR – Near infrared

Nm – Nanometer

NOCMC – N,O-carboxymethyl chitosan

NPs – Nanoparticles

O

O2 – Oxigen

OCMC – O-carboxymethyl chitosan

OX – Oxaliplatin

P

p – Statistical level of significance

Pa – Pascal

PAA - Polyacrylamide

PAMAM – Poly(amidoamine)

PAMAM-AT – poly(amidoamine)-amine terminated

PAMAM-CT – Poly(amidoamine)-carboxylic terminated

PBS – Phosphate buffer saline

PCL – Polycaprolactone

PDMS – Polydimethylsiloxane

PDX – Patient-derived xenograft

PEG – Poly (ethylene glycol)

xvii

PELA – Polyethylene glycol-co-polylactic acid

PGFs – Phosphate glass microfibers

pH – Potential hydrogenionic

PHBV – Polyhydroxybutyrate-co-hydroxyvalerate

PI – Propidium iodide

PL – Plantar muscles

PLA – Polylactic acid

PLCL – Poly(L-lactide-co-ε-caprolactone)

PLGA – Polylactic-co-glycolic

PLLA – Poly(L-lactic acid)

PVA -- Polyvinyl alcohol

% – Percentage

R

Ref. – Reference

RGD – Arginine-glycine-aspartic acid (Arg-Gly-Asp)

RM – Regenerative medicine

RNA – Ribonucleic acid

RFU – Relative fluorescence units

ROS – Reactive oxygen species

Rpm – Rotations per minute

RT – Room temperature

RT-PCR – Real time polymerase chain reaction

S

SD – Standard deviation

SDS – Sodium dodecyl sulfate

SEM – Scanning electron microscopy

SF – Silk fibroin

SiO2-NPs – Silica nanoparticles

SPIONs – Superparamagnetic iron oxide nanoparticles

ST – Short-term

T

xviii

TAF – Tumor associated fibroblasts

Tan δ – Loss factor

TCPs – Tissue culture polystyrene

TE – Tissue engineering

TERM – Tissue engineering and regenerative medicine

TG – Glass transition

TNF – Tumor necrosis factor

TRAIL – TNF-related apoptosis-inducing ligand

U

U – Units

UV – Ultraviolet

V

V – Volume

(v/v) – Percentage of volume/volume

VEGF – Vascular endothelial growth factor

X

XRD – X-ray diffraction

μg/mL – Microgram per milliliter

μL – Microliter

W

(w/v) – Percentage of weight/volume

WL – Weight loss

WU – Water uptake

%wt. – Percentage of weight

ɵ Water – Water contact angle

λ – Wavelength

Z

ZN – Zinc

ZnNPs – Zinc nanoparticles

xix

LIST OF EQUATIONS

Equation IV-1 – Encapsulation efficiency. .......................................................................................... 81

Equation IV-2 – Drug loading efficiency. ............................................................................................ 81

Equation IV-3 – Henry equation. ........................................................................................................ 99

Equation VII-1 – Determination of encapsulation efficiency. ............................................................. 199

Equation-VII-2 – Determination of drug loading efficiency. ................................................................ 199

xx

LIST OF FIGURES

Figure I-1 - Chip-Based 3D Models in Cancer Research. Scheme of an example of a chip-based 3D

model mimicking the native extracellular matrix (ECM) tumor microenvironment. The entry of

tumor cells into the blood stream and the micrometastasis process in a variety of tissues, or in the

presence of other relevant cell types, can be studied. The chip microwells could contain a 3D

engineered hydrogel representing an ECM-like matrix, such as Matrigel®, in direct contact with

microchannels mimicking blood vessels for intravasation and/or extravasation studies. The small

blue arrows represent the migration of (cancer) cells from the microchannel and into the 3D

tissue-engineered ECM-like matrix and vice-versa. (A) Cellix VenaT4chip; (B) The microwell filled

with Matrigel® (3D); (C) Microscopy image of the channel–microwell interface; (D) Representation

of cancer cells migrating from the microchannel to the 3D TE ECM-like matrix in the presence of

osteoblasts; (E) fluorescence microscopy image of HeLA cells adhered to the channel migrating

into the ECM-like matrix (Matrige®l) M. R. Carvalho, unpublished data, 2015). ......................... 12

Figure I-2 - Microfluidic devices used in the study of angiogenesis (A) and metastasis (B). A1) Passive

pumping-based microfluidic angiogenesis assay with 3D cylindrical lumens. (A) Illustration of a

triple channel design with connecting microchannels. (b–d) Microchannel systems can be (b)

single, (c) double, or (d) triple channel designs, and are arrayable. Reprinted with permission from

[95]. Copyright, 2012 Elsevier. a2) Perfusable three-dimensional microvessels are generated

using an optically clear polydimethsiloxane microfluidic-based platform. Reprinted with permission

from [94]. Copyright, 2013 BioMed Central Ltd. b) Microfluidic vasculature enables region-specific

activation of endothelium under physiological flow conditions [96]. ............................................ 13

Figure I-3 - Outputs of angiogenesis (A) and metastasis (B) studies in TE designs. A1) Cultured breast or

prostate epithelial carcinoma cells (MCF-7; LNCaP) with HUVECs and mesenchymal stromal cells

(MSCs) within matrices fabricated from synthetic starPEG and maleimide-functionalized heparin to

study 3D tumor angiogenesis microenvironments after 14 days: Extended focus confocal images

displaying Phalloidin (red), Hoechst (blue) and CD31 (green) showing HUVEC and MSC to cancer

cell interactions for each tumor cell type. Reprinted with permission from [104]. Copyright, 2015

Elsevier. A2) 3D endothelial-lined lumens (ELL) obtained by seeding Human umbilical vein

endothelial cells (HUVECs) in a microfluidic chip. Reprinted with permission from [95]. Copyright,

2012 Elsevier. B1) Microfluidic tumor-vascular interface model: Endothelial channel (green),

tumor channel (green) and 3D ECM (grey) between the two channels. White arrow shows

xxi

Fibrosarcoma cells (HT1080) invading in 3D toward the endothelium. [102]. B2) Migration of

HUVEC and Hela cells during co-culture on-chip. Reprinted with permission from [105]. Copyright,

2012 American Chemical Society. ............................................................................................ 15

Figure II-1 - Biomaterials applied in microfluidic systems. 1) In situ neuron patterning. Differentiated

human SH-SY5Y neurons cultured for 5 days within the CNA devices on a PL/PLL-g-PEG array of

micro-sized channels (A) and on a continuous PL coating (B). Biomaterial patterning increases

neurite guidance into the outgrowth microchannels. Reprinted with permission from [31].

Copyright © 2018 Royal Society of Chemistry. 2 - C) Optical image of hydrogel microstructure

arrays containing two phenotypes of cells: HepG2 and A549 cells. HepG2 cells were encapsulated

inside cylindrical hydrogel microstructures (78 μm Ø × 50 μm), while A549 cells were

encapsulated inside rectangular hydrogel microstructures (330 μm × 217 μm × 50 μm). D)

Corresponding fluorescence image for the cell viability expressed by live/dead assay (live cells

green, dead cells red). Reprinted with permission from [36]. Copyright © 2010 Elsevier. .......... 29

Figure II-2 - Microfluidics in drug discovery. 1) On-chip reaction. (A) Schematic of the on-chip reaction

process. (B) Time frames of a video that correspond to the steps described in A. A concentrated

solution of crystal violet (a pH indicator) in 0.1 N sodium hydroxide solution is loaded into the

microchamber with the three-state valve half-open; loading an equal amount of 1 N HCl into the

chamber turns the solution from blue to yellow. Reprinted with permission from [45]. Copyright ©

2004 The National Academy of Sciences; 2) Schematic diagram of reversible sealing of

microfluidic arrays onto microwell-patterned substrates to fabricate multiphenotype cell arrays.

Initially, a PDMS microfluidic mold in the form of an array of microchannels was aligned on an

array of microwells. Each cell type was flown through an independent channel. The cells docked

onto the corresponding micro-wells, which resulted in a patterned array of cells. To deliver

multiple solutions, the PDMS microfluidic mold was removed and replaced with another mold,

which was placed orthogonally to create multiphenotype cell arrays inside each microchannel.

Reprinted with permission from [54]. Copyright © 2018 Copyright Clearance Center, Inc. ......... 32

Figure II-3 - Organs-on-chip models: 1) A microengineered lung-on-a-chip model of human pulmonary

edema. (A) IL-2 therapy is associated with vascular leakage. (B) IL-2–induced pulmonary edema is

modeled in the lung-on-a-chip. The top or “air” portion is the alveolar channel; the bottom or

“liquid” portion is the vascular channel. Scale bar represents 200 μm. (C) Endothelial exposure to

IL-2 (1000 U/mL) causes liquid from the lower, microvascular channel to leak into the upper,

xxii

alveolar chamber. Scale bars represent 200 μm. (D) During IL-2 treatment, prothrombin (100

μg/mL) and fluorescently labeled fibrinogen (2 mg/mL) introduced into the microvascular channel

form fluorescent fibrin clots (white) over the course of 4 days. Scale bar represents 200 μm. (E) A

fluorescence confocal microscopic image shows that the fibrin deposits (red) in (D) are found on

the upper surface of the alveolar epithelium (green). Scale bar represents 50 μm. (F) The clots in

(D) and (E) are highly fibrous networks. Scale bar represents 5 μm. Reprinted with permission

from [56]. Copyright © 2012, American Association for the Advancement of Science. 2)

Microfluidic liver-on-a-chip model. (a) The liver receives blood flow from the hepatic artery and the

portal vein, which carries xenobiotics such as ingested drugs from the small intestine to be

metabolized. (b) Schematic of the hepatocyte microenvironment in liver tissue. (c) The sinusoid

space is bordered by a sheet of highly fenestrated endothelial cells. (d) The single microfluidic

sinusoid had three fluid terminals: a flow inlet, a flow outlet, and a cell inlet. Scale bar represents

20 μm. Reprinted with permission from [59]. Copyright © 2007 Wiley Periodicals, Inc. 3) Higher-

throughput heart-on–a-chip model. (a) Schematic of the fabrication process. (b) Image of the

engraving laser processing of Muscular Thin Films (MTFs) in the 18 mm diameter chip. (c)

Exploded view of the conception and assembly of the fluidic device which fits the chip. (d) Image

of an actual device in action. Reprinted with permission from [61]. Copyright © 2018 Royal

Society of Chemistry. 4) In vivo bone marrow-on-a-chip model. (a) Workflow to generate a bone

marrow-on-a-chip system (b) PDMS device containing bone-inducing materials in its central

cylindrical chamber before implantation (top). (c) Low- (left) and high-magnification views (right) of

histological hematoxylin-and-eosin–stained sections of the bone marrow formed in the PDMS

device with two openings (top) or one lower opening (center) at 8 weeks. Reprinted with

permission from [62]. Copyright © 2014, Springer Nature. ....................................................... 36

Figure III-1 - Rationale for nanomedicine approach in cancer-therapy. Reprinted with permission from

[1]. .......................................................................................................................................... 47

Figure III-2 - Dendrimer branched architecture, representing the increasing generations (G). Reprinted

with permission from [24]. ....................................................................................................... 49

Figure III-3 - Different types of dendrimers. Reprinted with permission from [33]. ............................... 51

Figure III-4 - Different strategies for employing dendrimer NPs in CRC research. A) Two synthetic

procedures of aSlex-conjugated dendrimers with FITC labeling by sequential conjugation of the

completely carboxylated PAMAM dendrimers with FITC and aSlex (a–c–e) or by sequential

xxiii

conjugation of the partially carboxylated PAMAM dendrimers with aSlex and IgG/IgM-FITC (b–d–f).

1, PAMAM with amine groups; 2, PAMAM with the partially carboxylated groups; 3, PAMAM with

the completely carboxylated groups; 4, PAMAM linked with FITC; 5, PAMAM conjugated with

primary antibody; 6, FITC-labeled PAMAM conjugated with antibody; 7, Primary antibody-coated

PAMAM conjugated with FITC-labeled secondary antibody. Reprinted with permission from [62].

Copyright © 2015 Nature. B) Dendrimer-gold hybrid structure synthesized by complexing AuCl4−

ions with PEGylated amine-terminated generation 5 PAMAM dendrimer. The resultant hybrid

system was loaded with curcumin. The curcumin-loaded PEGylated Au dendrimer was further

conjugated to MUC-1 aptamer in order to target the colorectal adenocarcinoma in vitro and in vivo.

Reprinted with permission from [12]. Copyright © 2018 Elsevier. C) Schematic representation of

(A) synthesis of pegylated PAMAMdendrimer (PEG-PAMAM); (B) camptothecin (CPT) loading in the

cavities of PEG-PAMAM; (C) conjugation of thiolated AS1411 aptamers to the maleimide groups of

MAL-PEG-PAMAM-CPT and preparation of Apt-PEG-PAMAM-CPT. Reprinted with permission from

[64]. Copyright © 2017 Elsevier. D) Synthetic route to polyoxazoline-modified dendrimers with

linker-SN38 conjugation. Reprinted with permission form [66]. Copyright © 2017 Elsevier. ....... 57

Figure IV-1 - Chemical structure of chitin and chitosan. Reprinted with permission from [5]. ............... 66

Figure IV-2 - Preparative methods of different types of Carboxymethylchitosan. Reprinted with

permission from [10]................................................................................................................ 67

Figure IV-3 – Matrigel® in its original packing and after cell encapsulation. ........................................ 72

Figure IV-4 – PDMS before and after crosslinking and plasma treatment (adhered to glass). .............. 74

Figure IV-5 - Schematic model of Laminin-111 showing the location of peptides that exhibit cell

attachment activity. Laminin-111 is composed of three subunits, α, β and ϒ chains. Forty-five

active peptides are localized in α1 chain, 14 active peptides in β1 chain and 12 active peptides in

ϒ1 chain. The four highlighted peptides described here are active in tumor malignancy. Reprinted

with permission from [60]. ....................................................................................................... 75

Figure IV-6 Structure of Laminin-related peptide TYR-ILE-GLY-SER-ARG. ................................. 76

Figure IV-7 - Schematic representation of dendrimer nanoparticles successive modifications with

carboxymethilChitosan and YIGSR peptide from laminin via carbodiimide chemistry. ................. 82

Figure IV-8 - Concentrated aqueous SF solution in a 50 mL Falcon. Scale bar: 1.5 cm. ...................... 84

xxiv

Figure IV-9 - Micropipettes Gilson M100 and M1000, specifically designed for hydrogel use. Scale bar: 1

cm. .......................................................................................................................................... 85

Figure IV-10 – Vena4™ empty and with silk hydrogels placed on the microwells. Scale bar: 6 mm. .... 86

Figure IV-11 - Simplified illustration of photolithography microfabrication. ........................................... 87

Figure IV-12 - Clewin 5 software design of microfluidic chip with close up on microchannel/microwell

interface. ................................................................................................................................. 88

Figure IV-13 – Example of serpentine like microchannels and stages of producing silk hydrogel

microfluidic platform. ............................................................................................................... 89

Figure IV-14 - µCP strategy applied in this thesis as proof-of-concept.................................................. 90

Figure IV-15 - A) Apparatus for generating gradients in real time. B) Microfluidic device under perfusion

to create a gradient. C) Fluorescence image of the FITC-CMCht/PAMAM nanoparticles’ gradient.

................................................................................................................................................ 92

Figure IV-16 - SEM with EDS analyzer. The SEM (JSM-6010 LV, JEOL, Japan) instrument is equipped

with the low vacuum mode as a standard feature. The low vacuum mode allows non-conductive

specimens to be observed and analyzed without conductive coating. This SEM is equipped with an

energy dispersive spectroscope (EDS). Scale bar: 10 cm. ......................................................... 93

Figure IV-17 - The Fourier transform infrared (FTIR) spectrometer is used as a standard chemical

characterization technique. The samples can be analyzed as films, KBr discs, powder or liquids.

The available methodologies comprise transmittance, specular reflectance, diffuse reflectance and

attenuated total reflectance (ATR). Scale bar: 10 cm. ................................................................ 94

Figure IV-18 - The AFM is able to acquire images of flat surfaces that can encode the surface

topography, mechanical response, among other properties. The equipment presents a series of

accessories that allows the acquisition of data in liquids and air. In the case of liquids, it is also

possible to control the temperature of the experiments. Scale bar: 10 cm. ................................ 96

Figure IV-19 - Differential scanning calorimetry is a thermal analysis technique that detects the

temperatures and heat flows caused by changes in heat capacity or by endothermic and

exothermic processes of materials as a function of time and temperature. Scale bar: 10 cm. .... 97

Figure IV-20 - Zetasizer Nano ZS, Malvern Instruments. Scale bar: 10cm. .......................................... 99

Figure IV-21 – Rheometer apparatus, Malvern Instruments. Scale bar: 10 cm. ................................ 102

xxv

Figure IV-22 – Schematics of Modified Boyden Chamber assay with the experimental conditions used in

Chapter VI. ............................................................................................................................. 109

Figure IV-23 - Scheme representation of endothelial cell invasion from lateral channels towards the

central chamber following VEGF gradient. Scale bar: 1 mm. .................................................... 111

Figure V-1 - Characterization of water soluble nanoparticles. A) Transmission electron microscopy (TEM)

image of the CMCht/PAMAM. B) Atomic force microscopy (AFM) images of CMCht/PAMAM on a

quartz mica. C) Particle size distribution (nm). D) H1NMR spectra of CMCht/PAMAM, PAMAM

G1.5 and CMCht. E) Differential Scanning Calorimetry (DSC) spectrum of CMCht/PAMAM. ..... 148

Figure V-2 - Normalization of MTS results by DNA results of HCT-116, HeLa and U87MG human cancer

cell lines in static conditions in the presence of CMCht/PAMAM dendrimer nanoparticles. Assays

were conducted to assess the cytotoxicity effect of nanoparticles on cell viability at concentrations

of 0.1, 0.5 and 1 mg/mL at time points of 1 and 3 days (* indicates significant differences when

comparing the different concentrations to control (0 mg/mL) at each time point. Ф indicates

significant differences when comparing controls from day 3 to day 1 to determine cell

proliferation). ......................................................................................................................... 149

Figure V-3 - Cytotoxicity of CMCht/PAMAM over HCT-116, HeLa and U87MG human cancer cell lines in

static and dynamic conditions. Assays were conducted to assess cytotoxicity/viability over the cells

at a concentration of 0.5 mg/mL at different time points (days 1 and 3) and it in static and

dynamic conditions. (* indicates significant differences when comparing 0.5 mg/mL to control (0

mg/mL) at each time point. Ф indicates significant differences when comparing controls from day

3 to day 1 to determine cell proliferation). ............................................................................... 150

Figure V-4 - Fluorescent microscope images of FITC-labeled CMCht/PAMAM dendrimer nanoparticles

(green) internalization in cancer cells labelled with DAPI (nuclei) and Phalloidin (F-actin,

cytoskeleton) after 1 and 3 days in direct contact with dendrimer nanoparticles in static

conditions. ............................................................................................................................. 151

Figure V-5 - Fluorescent microscope images of FITC-labeled CMCht/PAMAM dendrimer nanoparticles

(green) internalization in cancer cells labeled with DAPI (nuclei) and Phalloidin (F-actin,

cytoskeleton) after 24 hours and 72 hours in direct contact in dynamic conditions. ................. 153

Figure VI-1 - Quantification of outward migration of HCT-116 (labeled with Red Cell Tracker in modified

Boyden chamber from 2 and 3 % SF hydrogels and Matrigel coating, in response to the presence

xxvi

of hVCAM-1 and 30 % (v/v) of FBS. (* indicates significant differences when comparing to Matrigel

coating at each time point; • indicates significant differences when comparing to 2 % SF at each

time point; $ indicates significant differences when comparing to VCAM at each time point; ᶲ

indicates significant differences when comparing to FBS at each time point; # indicates significant

differences when comparing to time point 2 hours. ................................................................. 175

Figure VI-2 - Representative panel of outward migration of HCT-116 cells from 2 % and 3 % SF and

Matrigel coating in response to gradients of hVCAM-1 and FBS. Migrating cells attached to the

lower side of the insert membrane at 48 hours stained with DAPI (blue) and Phalloidin (Red). The

first column represents the pores of the membrane without cells. N/A denotes no cell migration.

.............................................................................................................................................. 177

Figure VI-3 - A) Example of Vena4™ chip with SF hydrogels placed on the microwells. B) Vena4™ placed

under the confocal microscope for migration studies. C) Retrieval of SF hydrogels out of the

microfluidic chip for further analysis ....................................................................................... 178

Figure VI-4 - Representative confocal microscopy images of red-labeled HCT-116 cells’ migration

towards 3 % and 2 % SF hydrogels and Matrigel, (with and without hVCAM-1) in microfluidic chip

Vena4™ at 48 hours. .............................................................................................................. 180

Figure VI-5 - A) Cell viability (live/dead assay) of HCT-116 cells that migrated in the hydrogels after 48

hours of culture in Vena4™ microfluidic chip. B) Cell viability of 3D cultures after 48 hours of

culture (2 % and 3 % SF and Matrigel hydrogels with hVCAM-1). Data is presented as mean ±

stdev (n=3), (*) denotes statistical differences (p<0.05). ......................................................... 182

Figure VII-1 - Design and characterization of the microfluidic chip. A) Profilometer characterization: 3D

map of the microfluidic chip. Representative image of quality control and characterization of the

microfluidic chip features. The central chamber is 5 mm in diameter and 126 µm in depth, with a

separate inlet and outlet for hydrogel injection. The two lateral channels, each 100 µm wide and

126 µm deep, are not interconnected, opening up the possibility of perfusing two distinct

solutions. B) Schematic of chip design and zoom in of the concept of colorectal tumor-on-a-chip

model: round microfluidic central chamber in which HCT-116 cancer cells are embedded in

Matrigel; HCoMECs are seeded in the side channels to form a 3D vessel-like assembly. C)

Establishment of a microvascular 3D microenvironment of colorectal tumor-on-a-chip: HTC-116

CRC cells embedded in Matrigel (in central chamber, stained in red) and HCoMECs (in lateral

microchannels, stained in yellow) at day 1 and 5 of culture. Endothelial cells start to invade the

xxvii

central chamber filled with Matrigel in response to VEGF presence. D) Fluorescence microscopy

image of microfluidic lateral channel mimicking pre-vascularization with HCoMECs cells after 5

days of culture (DAPI blue – Nuclei; Phalloidin green – F-actin filaments); Representative

fluorescence microscopy close-up image of microchannel cross section showing endothelial cells

aligning and creating an endothelialized lumen within the microchannel. ................................. 204

Figure VII-2 - Endothelial cell invasion. A) Schematic representation of HCoMECS cells in the lateral

channels invading the central chamber in response to the presence of VEGF mixed in the Matrigel;

B) Formation of endothelial sprouts in the microfluidic device: representative bright field images of

the chip taken at determined time points were analyzed in Image J. C) Quantification of

endothelial invasion. Data is presented as mean ± stdev (n=9), (*) denotes statistical differences

(p < 0.01). ............................................................................................................................. 205

Figure VII-3 - Schematic representation of fluorescent nanoparticle gradient formation. A) Transmission

electron microscopy image of individual CMCht/PAMAM dendrimer nanoparticle, representative

image of chip connected to tubing perfusing the nanoparticle solution, and zoom in confocal

microscopy image confirming FITC-labeled CMCht/PAMAM dendrimer nanoparticles’ dispersion in

Matrigel. B) Panel of fluorescent microscope images of FITC-labeled nanoparticles’ gradient up to

12 hours. C) Quantification of fluorescence: intensity (arbitrary units) vs distance (µm) across the

chip up to 12 hours (measurements according to the direction of the green arrow indicated in B).

D) 3D projection of FITC-labeled nanoparticles’ gradient according to fluorescence intensity. ... 206

Figure VII-4 - GEM drug effect on colorectal-tumor-model. A) Release profile of GEM: cumulative release

(hours) of GEM at pH 7.4 in PBS, at 37 ºC and stirred at 60 rpm, determined by UV

spectrophotometer set at 275 nm. The results are expressed as mean ± stdev, n=6. B) Schematic

representation of experimental set-up (perfusion of culture medium through one in- and one outlet

and of culture medium supplemented with 0.5 mg/mL GEM-loaded dendrimer nanoparticles

through the other in- and outlet); the areas defined for cell death quantification are delimited by

dashed lines and each ring assigned D1-D3 and M1-M3. C) Representative panels of fluorescence

microscopic images of live/dead assay performed on microfluidic chip at days 1 and 5. D)

Quantification of cell death, represented as percentage, based on the microscopy data. Data is

presented as mean ± stdev (n=3), (*) denotes statistical differences (p < 0.05). Scale bar 1000

µm. ....................................................................................................................................... 209

xxviii

Figure VII-5 - Gene expression analysis and immunocytochemistry on a chip. A) Representation of

hydrogel retrieval containing HCT-116 cells from the chip for gene expression analysis. B) Gene

expression of Ki-67, Caspase-3 and MMP-1, in the absence (control) and presence of gemcitabine,

at day 1 and day 5. The values for controls are represented by the dashed line. * Significant

differences when comparing the same condition at two different time points; # significant

differences when comparing to controls at the same time point. Data represent mean values of 3

independent experiments ± stdev. C) Immunohistochemistry: in the center, a representative image

of the entire chip is depicted. The lateral images represent the zoom in at the cell level showing

the immunocytochemistry Ki-67 staining after 5 days in culture (panels showing DAPI nuclei

staining blue, Ki-67 staining green, F-actin staining red and merged image in colors). .............. 211

Figure VIII-1 - Physico-chemical characterization of dendrimer nanoparticles. (A) Schematic

representation of the dendrimer NPs modification with YIGSR peptide from laminin via

carbodiimide chemistry. (B) STEM image of CMCht/PAMAM and YIGSR-CMCht/PAMAM

dendrimer NPs. (C) Representative graphic of the particle size distribution of CMCht/PAMAM and

YIGSR-CMCht/PAMAM NPs, showing that the peptide sequence conferred the NPs a larger size.

(D) FTIR analysis of CMCht/PAMAM and YIGSR-CMCht/PAMAM dendrimer NPs, showing an

increase in intensity of the bands at 1700 cm-1 for YIGSR-CMCht/PAMAM dendrimers. (E) H1NMR

spectra of YIGSR-CMCht/PAMAM, CMCht/PAMAM, and YIGSR. The YIGSR-CMCht/PAMAM shows

a peak at 3 ppm and no peak at 4.3 ppm suggesting that the substitutions occurred mainly on the

C2 amino group of CMCht. (F) Table of the mean NP diameter and Zeta potential in PBS and

ddH20, showing an increase in diameter and a decrease in Zeta potential upon the addition of

YIGSR. (G) Determination of peptide concentration by protein quantification (BCA protein

quantification) in CMCht/PAMAM and YIGSR-CMCht/PAMAM. ................................................ 234

Figure VIII-2 - Specificity of FITC-YIGSR-CMCht/PAMAM dendrimer NPs on a micropatterned LR array.

(A) Immunocytochemistry of anti-67kDa LR (green) performed on a co-culture of HCT-116 CRC

cells and L929 fibroblasts (stained in red; See supplementary Information S3) cells at 24 h and

72 h. An overexpression of LR on HCT-116 is visible at both time points. (B) (Left) Schematic

representation of the steps involved in the microcontact printing of the LR: (i) incubation of the

PDMS stamp with the LR solution; (ii) microcontact printing of the LR on top of a functionalized

(APTES 5 % and glutaraldehyde) glass coverslip; (iii) PDMS stamp removal and rinsing with ddH2O;

(iv) incubation with FITC-YIGSR-CMCht/PAMAM; and (v) rinse with ddH2O to remove the excess of

FITC-YIGSR-CMCht/PAMAM. (Right) Representative brightfield and fluorescence microscopy

xxix

images of the PDMS stamp and FITC-YIGSR-CMCht/PAMAM NPs array displaying the triangular

micro-sized LR features. Scale bar: 100 μm. .......................................................................... 236

Figure VIII-3 - Cytotoxicity studies. (A) Metabolic activity determined by Alamar blue, and (B) DNA

quantification results of HCT-116 and L929 cells in the presence of 0.5 mg/mL of

CMCht/PAMAM, YIGSR-CMCht/PAMAM, and GEM-loaded YIGSR-CMCht/PAMAM, at 24 hours, 48

hours and 72 hours. *Significant differences when comparing the different conditions with control

(0 mg/mL) at each time point. # significant differences were observed when comparing to

CMCht/PAMAM at each time point. $ Significant differences were observed when comparing to

YIGSR-CMCht/PAMAM at each time point. P < 0.05. .............................................................. 237

Figure VIII-4 - Internalization of dendrimer nanoparticles. A) Schematics of the different conditions

studied. B) Confocal microscopy images of HCT-116 cancer cells (red) and L929 fibroblasts (blue)

co-culture in the presence of YIGSR-CMCht/PAMAM and CMCht/PAMAM dendrimer NPs in (top)

2D for 24 hours and 72 hours, and (bottom) in 3D for 72 hours. In the latter, (1) and (2) show a

higher magnification of HCT-116 and L929 cells, respectively, showing the internalization of

YIGSR-CMCht/PAMAM dendrimer NPs exclusively for HCT-116 (observed only with the 488 nm

filter). Inset images show a more detailed view of the NPs (indicated by white arrowheads). (C)

Flow cytometry results of NPs internalization in co-culture in both 2D and 3D for 24 hours and 72

hours. Scale bars: 20 µm. ...................................................................................................... 240

Figure VIII-5 - The cell viabilities of HCT-116 and L929 cells in the presence of GEM-YIGSR-

CMCht/PAMAM dendrimers. (A) Live/Dead assay performed on a co-culture of HCT-116

(unstained) and L929 (in blue) cells at 24 hours and 72 hours in 2D. (B) Flow cytometry analysis

of cell viabilities in the presence of GEM-YIGSR-CMCht/PAMAM in a co-culture of HCT-116 (CD44

positive) and L929 cells (CD44 negative cells). ....................................................................... 243

Figure IX-1 - Illustration the process of producing the PDMS mold used to produce the enzymatically

crosslinked silk fibroin hydrogel-based microfluidic device. ...................................................... 265

Figure IX-2 – ATR-FTIR spectra for the enzymatically crosslinked silk fibroin hydrogel (with 12 %

concentration of silk fibroin) retaining the amorphous protein structure for at least 7 days. ...... 265

Figure IX-3 – Demonstration of the flexibility and elasticity of the transparent enzymatically crosslinked

silk fibroin hydrogel at day 1 and at day 3. .............................................................................. 266

xxx

Figure IX-4 - Schematically illustrates an exemplification of the structure of the enzymatically crosslinked

silk fibroin microfluidic device. In this case, including a meandering serpentine channel, inlets and

outlets. Figure 4A is without dimensions while Figure 4B is dimensions in µm. ........................ 266

Figure IX-5 - SEM images of the 3D enzymatically crosslinked silk fibroin hydrogel structures after drying

using point drying. Scale bars: 500 µm (a and b), and 100 µm (c). ......................................... 267

Figure IX-6 - Schematically illustration of an example of what the enzymatically crosslinked silk fibroin

hydrogel can be used for. ....................................................................................................... 267

Figure IX-7 - Confocal microscopy images as described in the schematics of figure 6: endothelial cells

seeded on the microchannel while colorectal cells are encapsulated throughout the enzymatically

crosslinked silk fibroin microfluidic device. .............................................................................. 268

Figure IX-8 - Viability of cells encapsulated in the enzymatically crosslinked silk fibroin hydrogel-based

microfluidic device. Live/dead assay was performed and observed under confocal microscope.

.............................................................................................................................................. 268

Figure IX-9 - Perfusion of blue ink through the inlet (A); formation of soluble food colouring ink gradient

(B); and diffusion of ink visible in the microchannels, allowing for the formation of diffusion

gradients of drugs/nanoparticles. ........................................................................................... 269

Figure IX-10 - Liquid perfusion through the enzymatically crosslinked silk fibroin hydrogel-based

microfluidic device’s microchannel. (A) Shows perfusion of blue ink through the inlet. (B1) and

(B2) show magnified images of liquid perfusion inside the serpentine microchannel (or

meandering section). .............................................................................................................. 269

xxxi

LIST OF SUPPLEMENTARY FIGURES

Supplementary Figure VIII-1 - Nanoparticles characterization. (A) AFM images of the CMCht/PAMAM

and YIGSR-CMCht/PAMAM NPs with respective 3D projection; (B) Peptide determination by UV

spectroscopy; (C) Flow Cytometry graphic showing selective binding of CD44 antibody to HCT-116

in a co-culture suspension with L929 cells; (D) SEM images of a co-culture of HCT-116 cells and

L929 cells: close ups showing the nanoparticles do not agglomerate on the cell’s surface, but are

widely internalized instead; and (E) Confocal image of HCT-116 (red) and L929 cells (blue) in 3D,

in the presence of FITC-YIGSR-CMCht/PAMAM NPs at 24 hours. ............................................ 246

Supplementary Figure VIII-2 - H1NMR analysis of YIGSR peptide, CMCht/PAMAM, and YIGSR-

CMCht/PAMAM. A) H1NMR (D2O, 400 MHz) spectra of the peptide YIGSR with its chemical

structure and main peaks. B) H1NMR (D2O, 400 MHz) spectra of CMCht/PAMAM with peak

assignments. C) H1NMR (D2O, 400 MHz) spectra of the modified YIGSR-CMCht/PAMAM and

respective peak assignment. D) H1NMR (D2O, 400 MHz) spectra of: the modified YIGSR-

CMCht/PAMAM (D1); CMCht/PAMAM dendrimer (D2), and theYIGSR (D3), with arrows indicating

the main peaks with the peaks marked the color according to the respective provenance. ....... 247

Supplementary Figure VIII-3 - Immunocytochemistry of anti-67kDa LR (green). Pannel A shows 488 filter

(green) for the 67 kDa receptor, and panel B shows 594 filter (red) performed on a co-culture of

HCT-116 CRC cells and L929 fibroblasts ( previsouly stained in red) cells at 72 hours, proving that

the cluster of HCT-116 overexpressing 67 kDa receptor consists in fact of cancer cells, and not

L929. ..................................................................................................................................... 247

xxxii

LIST OF TABLES

Table I-1 - Naturally Derived Matrices for 3D Tumor Engineering. ......................................................... 6

Table III-1 - Current nanotechnology applications in colorectal cancer clinical trials and status. ........... 47

Table III-2 - Summary of different strategies using dendrimer NP for CRC therapy. ............................. 56

Table IV-1 - Comparative data on amino acid composition of B. mori, N. edulis and N. clavipes silks.

Reprinted with permission from [15]. ........................................................................................ 68

Table IV-2 - Primers list for the studied genes. ................................................................................. 122

Table IV-3 – Antibodies list used in immunocytochemistry. .............................................................. 124

Table V-1 - FACS data of the percentage of internalization of FITC-labeled CMCht/PAMAM dendrimer

nanoparticles (NP) and the percentage of live cells (stained with 7AAD), from day 1 to day 3 in

static conditions. .................................................................................................................... 152

Table V-2 - FACS data of the percentage of internalization of FITC-labeled CMCht/PAMAM dendrimer

nanoparticles and the percentage of live cells (stained with 7AAD), from day 1 to day 3 in dynamic

conditions. ............................................................................................................................. 154

Table VI-1 – Composition, rheological properties and original mesh size of hydrogels at a frequency of

0.1 Hz. .................................................................................................................................. 174

Table VIVII-1 – Primers list for the studied genes. ............................................................................ 202

Table IX-1 - Composition, rheological properties of hydrogels at a frequency of 0.1 Hz. .................... 260

xxxiii

SHORT CURRICULUM VITAE

Mariana Rodrigues de Carvalho was born in 1988 in Guimarães, Portugal. Nowadays, she

works as a researcher in 3B's Research Group, I3Bs – Research Institute on Biomaterials,

Biodegradables and Biomimetics, University of Minho, under the supervision of Prof. Rui L. Reis and Dr.

Joaquim M. Oliveira.

Her background includes a five-year graduation (integrated masters) in Pharmaceutical

Sciences at the Faculty of Pharmacy of the University of Porto finished in 2012. During the graduation

she enrolled in research activities resulting in authorship of a paper entitled “Teas, dietary supplements

and fruit juices: A comparative study regarding antioxidant activity and bioactive compounds”. Also, in

her last year, she had the opportunity to do a three-month internship at the World Health Organization

Headquarters in Geneva, Switzerland. That gave her great insights into the Public Health field and the

opportunity to work closely with global leaders in public health.

In the same year, she joined 3B’s Research Group, after being awarded a grant under the

scope of Fundação para a Ciência e Tecnologia (FCT) project OSTEOCART, under the supervision of

Prof. Rui L. Reis and Dr. Miguel Oliveira. Since January of 2014, she continued her work at 3B’s

Research Group under the support of the FCT project PEst - 2013-2014, and European Commission

(EC) founded project NOVOMAR, with supervision of Prof. Rui L. Reis and Dr. Joaquim M. Oliveira.

During this period, she focused on studying several applications of CMCht/PAMAM dendrimer

nanoparticles as well as microfluidic technologies applied in cancer research. Mariana Rodrigues de

Carvalho has been involved in the preparation of FCT project proposals as well as European project

proposals, related with her work. Additionally, she collaborated with her colleagues in different works,

and different universities, including Medical Faculty of University of Zagreb and MERLN Institute at the

University of Maastricht.

As a result of her research work, she is author or co-author of 8 papers in international journals

(all published, including one published in Science Advances), 3 book chapters and 1 issued patent, 10

oral presentations and 9 poster presentations. She attended several important international meetings in

the field of tissue engineering and regenerative medicine. Her work was constituted a press release in

several occasions, by Associação Portuguesa de Investigação em Cancro (ASPIC), as well as the

University of Minho official website. One of her original papers was also featured as the cover of the

journal Global Challenges.

xxxiv

LIST OF PUBLICATIONS

The work performed during the PhD period resulted in the publications listed below.

Papers in international scientific journals with referees (as first author)

1. M. R. Carvalho, D. Lima, R. L. Reis, V. M. Correlo, J. M. Oliveira. “Evaluating biomaterials

and microfluidics based 3D tumor models”. Trends in Biotechnology, 2015, 33:667-678.

doi: 10.1016/j.tibtech.2015.09.009.

2. M. R. Carvalho, D. Lima, R. L. Reis, J. M. Oliveira, V. M. Correlo. “Anti-cancer drug

validation: the contribution of tissue engineered models”, Stem Cells Reports and Reviews,

2017, 3:347-363. doi: 10.1007/s12015-017-9720-x.

3. M. R. Carvalho, R. L. Reis, J. M. Oliveira. “Mimicking the 3D biology of osteochondral

tissue with microfluidic-based solutions: breakthroughs towards boosting drug testing and

discovery”. Drug Discovery Today, 2017, 3: 711-718. doi:

10.1016/j.drudis.2018.01.008.

4. M. R. Carvalho, F. R. Maia, J. Silva-Correia, B. M. Costa, R. L. Reis, J. M. Oliveira. “A semi-

automated microfluidic platform for real-time tracking of cancer cells and investigation of

nanoparticles cellular uptake”. Nanomedicine, 2016, 2:581-596. doi: 10.2217/nnm-

2016-0344.

5. M. R. Carvalho, F. R. Maia, S. Vieira, R. L. Reis, J. M. Oliveira. “Tuning Enzymatically-

crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer

Extravasation 3D Model on a Chip”. Global Challenges, 2018, 2: 1870164. doi:

10.1002/gch2.201700100.

6. M. R. Carvalho, D. Barata, L. M.Teixeira, S. Giselbrecht, R. L. Reis, J. M. Oliveira, R.

Truckenmüller, P. Habibovic. “Colorectal tumor-on-a-chip system: a 3D tool for precision

onco-nanomedicine”. Science Advances, 2019, 5: eaaw1317. doi:

10.1126/sciadv.aaw1317.

7. M. R. Carvalho, C. R. Carvalho, F.R. Maia, D. Caballero, S. C. Kundu, R. L. Reis, J. M.

Oliveira. “Peptide-modified dendrimer nanoparticles for targeted therapy of colorectal

cancer”. Advanced Therapeutics, 2019, 1900132. doi: 10.1002/adtp.201900132.

xxxv

Cover pictures in international scientific journals with referees

1. M. R. Carvalho, F. R. Maia, S. Vieira, R. L. Reis, J. M. Oliveira. Tuning Enzymatically-

crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer

Extravasation 3D Model on a Chip. Global Challenges, Vol. 2, No. 5/6, Published June 26,

2018.

Papers in international scientific journals with referees (as co-author)

1. S. Vieira, S. Vial, F. R. Maia, M. R. Carvalho, P. L. Granja, J. M. Oliveira. “Gellan Gum-

coated Gold Nanorods: An intracellular nanosystem for bone tissue engineering”. RSC

Advances, 2015, 5: 77996-78005. doi: 10.1039/C5RA13556G.

Book Chapters

1. M. R. Carvalho, F. R. Maia, J. M. Oliveira, R. L. Reis. Advanced microfluidics for 3D biology

in Osteochondral Tissue Engineering in Osteochondral Tissue Engineering -

Nanotechnology, Scaffolding-Related Developments and Translation, Advances in

Experimental Medicine and Biology, JM Oliveira, S Pina, RL Reis and JS Roman (Eds),

Springer International Publishing AG, 2018, ISBN: 978-3-319-76710-9

2. M. R. Carvalho, R. Truckenmuller, R. L. Reis J. M. Oliveira, Biomaterial- and Microfluidic-

Based 3D Models, Springer, Ed: J. Miguel Oliveira and Rui L. Reis (2019).

3. F. R. Maia, M. R. Carvalho, J. M. Oliveira and R. L. Reis. Tissue Engineering Strategies for

Osteochondral Repair in Osteochondral Tissue Engineering - Challenges, Current

Strategies, and Technological Advances, JM Oliveira, S Pina, RL Reis and JS Roman (Eds),

S. Springer, 2018, ISBN: 978-3-319-76734-5.

Conference oral presentations

1. M. R. Carvalho, J. Mano, R. L. Reis, J. M. Oliveira. Investigation of Dendrimer-based

nanoparticles cellular uptake and cell tracking in a semi-automated microfluidic platform.

Society for Biomaterials 2015 – North Carolina, USA, 15th-18th April, 2015.

2. M. R. Carvalho, F. R. Maia, R. L. Reis, J. M. Oliveira. Investigation of Dendrimer-based

nanoparticles cellular uptake and cell tracking in a semi-automated microfluidic platform.

European Conference on Biomaterials (ESB2015), Krakow, Poland, 30th August to 3rd

September, 2015.

xxxvi

3. S. Vial, S. Vieira, F. R. Maia, M. R. Carvalho, R. L. Reis, P. L. Granja, J. M. Oliviera. Gellan-

Gum Coated Gold Nanorods as Intracellular Drug Release System for Osteogenic

differentiation. European Conference on Biomaterials (ESB2015), Krakow, Poland, 30th

August to 3rd September, 2015.

4. M. R. Carvalho, F.R. Maia, J. Silva-Correia, B. M. Costa, R.L. Reis, J. M. Oliveira. A

microfluidic platform for the investigation of nanoparticles cellular uptake and cancer cells

tracking in real-time. TermStem 2016, Guimarães, Portugal, 24th-26th October, 2016.

5. M. R. Carvalho, F.R. Maia, J. Silva-Correia, B. M. Costa, R.L. Reis, J. M. Oliveira. A semi-

automated microfluidic platform for real-time tracking of cancer cells and investigation of

nanoparticles cellular uptake. Society for Biomaterials 2017, Minneapolis, USA, 4th-8th April,

2017.

6. M. R. Carvalho, F. R. Maia, J. Silva-Correia, B. M. Costa, R. L. Reis, J. M. Oliveira. A

microfluidic platform for the investigation of cancer cells’ nanoparticle uptake and real

time tracking. Chem2Nature Second School, Porto, Portugal, 5th to 9th of June, 2017.

7. M. R. Carvalho, D. Barata, L. M. Teixeira, S. Giselbrecht, R. L. Reis, J. M. Oliveira, R.

Truckenmüller, P. Habivović. Engineering a microvascular 3D microenvironment of

colorectal tumor-on-a-chip. The Netherlands Society for Biomaterials and Tissue

Engineering (NBTE) 26th Annual Meeting, Luteren, Netherlands, 3rd-6th April, 2017.

8. M. R. Carvalho, D. Barata, L. M. Teixeira, S. Giselbrecht, R. L. Reis, J. M. Oliveira, R.

Truckenmüller, P. Habivović. Engineering a microvascular 3D microenvironment of

colorectal tumor-on-a-chip. MicroNano Conference, Amsterdam, Netherlands, 11th-12th

December, 2018.

9. M. R. Carvalho, D. Barata, L. M. Teixeira, S. Giselbrecht, R. L. Reis, J. M. Oliveira, R.

Truckenmüller, P. Habivović. Microfluidic gradient generator for drug testing on a

colorectal tumor-on-a-chip disease model. Society for Biomaterials 2019, Seattle, USA, 4th-

8th April, 2017.

10. M. R. Carvalho, C. Carvalho, F.R. Maia, D. Caballero, S. C. Kundu, R. L. Reis, J. M.

Oliveira Targeting colorectal cancer using highly specific peptide-modified CMCht/PAMAM

dendrimer nanoparticles. TERMIS-EU, Rhodes, Greece, 27th-31st May, 2019.

xxxvii

Conference posters

1. M. R. Carvalho, J. Mano, R. L. Reis, J. M. Oliveira. Dendrimer based Nanoparticles for

intracellular delivery and cell tracking. TERMSTEM 2014 PT-Korea Symposium, Porto,

Portugal, 23th and 24th October, 2014.

2. M. R. Carvalho, J. Mano, R. L. Reis, J. M. Oliveira. Novel Dendrimer nanoparticles for iPS

generation in bone and cartilage regeneration. 3rd ICVS/3B’s - Associate Laboratory

Meeting, Braga, Portugal, 28th June, 2014.

3. S.Vial, S. Vieira, M. R. Carvalho, R. L. Reis, P. L. Granja, J. M. Oliveira. Gellan-Gum Coated

Gold Nanorods as Intracellular Drug Release System for Regenerative Medicine. Society for

Biomaterials, North Carolina, USA, 15th-18th April, 2015.

4. M. R. Carvalho, F. R. Maia, R. L. Reis, J. M. Oliveira. Investigation of Dendrimer-based

nanoparticles cellular uptake and cell tracking in a semi-automated microfluidic platform.

10th World Biomaterial Congress, Montreal, Canada, May 17th to 22nd, 2016.

5. M. R. Carvalho, F. R. Maia, J. Silva-Correia, B. M. Costa, R. L. Reis, J. M. Oliveira. A

microfluidic platform for the investigation of cancer cells’ nanoparticle uptake and real

time tracking. TERMIS-EU Meeting, Davos, Switzerland, 26th to 30th June, 2017. 20172013,

DOI:10.1002/term.1822

6. M. R. Carvalho, D. Barata, L. M. Teixeira, S. Giselbrecht, R. L. Reis, J. M. Oliveira, R.

Truckenmüller, P. Habivovic. Engineering a microvascular 3D microenvironment of

colorectal tumor-on-a-chip. The Netherlands Society for Biomaterials and Tissue

Engineering (NBTE) 26th Annual Meeting, Luteren, Netherlands, 28th-29th November,

2017.

7. M. R. Carvalho, F. R. Maia, S. Vieira, R. L. Reis, J. M. Oliveira. Tuning Enzymatically-

crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer

Extravasation 3D Model on a Chip. Chem2nature Summer School, Porto, Portugal, June

4th-8th, 2018.

8. M. R. Carvalho, D. Barata, L. M. Teixeira, S. Giselbrecht, R. L. Reis, J. M. Oliveira, R.

Truckenmüller, P. Habivovic. Engineering a microvascular 3D microenvironment of a

colorectal tumor-on-chip. ESB 2018, Maastricht, Netherlands, September 9th to 13th, 2018.

xxxviii

9. M. R. Carvalho, D. Barata, L. M. Teixeira, S. Giselbrecht, R. L. Reis, J. M. Oliveira, R.

Truckenmüller, P. Habivovic. FORECAST Workshop, Porto, Portugal, June 8th-12th 2019.

Patents

1. M. R. Carvalho, D. Caballero, C. R. Carvalho, J. B. Costa, V. Ribeiro, Oliveira J. M., Kundu

S. C., Reis R. L. Enzymatically Crosslinked Silk Fibroin hydrogel microfluidic platform,

methods of production and uses thereof. (Filed, 2019).

Awarded grants

1. Horizonte Norte2020 PhD scholarship (Norte-08-5369-FSE-000044).

2. EMBO Short-Term Fellowship 7232.

Press Release

1. “Combinação de nanopartículas e dispositivos de microfluidos promete impulsionar a

investigação sobre o cancro e terapias personalizadas”, ASPIC, 15th March 2017

(http://www.aspic.pt/pt-pt/noticias/combina%C3%A7%C3%A3o-de-nanopart%C3%ADculas-e-

dispositivos-de-microfluidos-promete-impulsionar#.W5E9pehKg2x).

2. “Grupo 3B’s da UMinho desenvolve teste para medicamentos contra o cancro colorretal. O

trabalho foi publicado na Science Advances e é “um avanço importante na nanomedicina de

precisão”, University of Minho Official website, July 25th 2019.

(https://www.facebook.com/search/top/?q=universidade%20do%20minho%20-

%20oficial&epa=SEARCH_BOX).

3. “Grupo 3Bs desenvolve teste para medicamentos do cancro colorectal”, Reflexo Digital

Official Website, July 22nd 2019. (https://reflexodigital.com/grupo-3bs-desenvolve-teste-para-

medicamentos-do-cancro-colorectal/)

xxxix

INTRODUCTION TO THE THESIS FORMAT

The present thesis is divided into four main sections (1 to 4) containing ten chapters (I to X).

This structure was adopted to allow a proper organization of the data presented in the various chapters.

A general introduction (Section 1) can be found divided in three different chapters: Chapters I, II

and III. An overall Materials and Methods section is represented in Section 2, Chapter IV. Section 3, in

which Chapters V to IX can be found, focuses on the experimental studies and results obtained in the

context of this thesis and their discussion. To finalize, Section 4 (Chapter X) completes this thesis with

concluding remarks and future perspectives on the topic.

The main body of the thesis is based on a series of publications published in international

journals or already submitted for publication. Each individual chapter is presented in a manuscript form,

i.e. abstract, introduction, experimental section, results and discussion, conclusion, and

acknowledgements. A list of relevant references is also provided as a subsection within each chapter.

The contents of each part and chapter are described below in more detail.

Section 1 – General introduction

Chapter I – Evaluating Biomaterial and Microfluidic-Based 3D Tumor Models: This chapter

presents an extensive overview on the development and combination of microfluidic and biomaterials

for cancer research, as well as the several aspects that make this technology suitable for promising

advances in cancer field, namely on angiogenesis and metastasis processes.

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development: This chapter

reviews the current reports dealing with the combination of biomaterials and microfluidics for drug

discovery in organ-on-chip, with special emphasis on tumor-on-a-chip.

Chapter III – Dendrimer Nanoparticles in CRC Applications: This chapter focuses on dendrimer

nanoparticles, types, routes of synthesis, and main applications in cancer research (with emphasis in

colorectal cancer research).

Section 2 – Detailed description of experimental materials and methodologies

Chapter IV – Materials and Methods: A list of the materials used and methods applied to obtain

the results described further on is provided, being the basis to the whole work described in this thesis.

Section 3 – Experimental work regarding the biomaterials approach to cancer in vitro models

on a chip

xl

Chapter V – A Semi-automated Microfluidic Platform for Real-time Tracking of Cancer Cells and

Investigation of Nanoparticles Cellular Uptake: In this chapter, several cancer cell lines were cultured in

the presence of dendrimer nanoparticles, and its effects compared in dynamic (microfluidic platform)

and static (traditional 2D culture flasks) conditions. Proof of the relevance of physiological aspect was

achieved.

Chapter VI – Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the

Development of a Colorectal cancer Extravasation 3D Model on a Chip: This chapter focuses on the use

of enzymatically crosslinked silk fibroin hydrogels with tunable mechanical properties to assess

colorectal cancer cells’ migration in response to the presence of hVCAM-1. Moreover, this chapter

focuses on the fundamental differences between the traditional Boyden Chamber migration assay and a

3D microfluidic migration assay.

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D tool for Precision Onco-nanomedicine:

In this chapter, we move forward with the goal of creating an in-house tailored made microfluidic device

for the development of an in vivo like model of colorectal cancer. This model comprises HCT-116

colorectal cancer cell embedded in Matrigel, connected to pair of laterally perfused channels with

lumen-creating colonic microvascular endothelial cells. The effect of a gradient of gemcitabine released

from dendrimer nanoparticles was evaluated, as well as gene expression.

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal

Cancer: In this chapter, it is described the use of CMCht/PAMAM dendrimer nanoparticles

functionalized with YIGSR laminin receptor binding peptide for actively targeting and specific delivery of

therapeutic agents into colorectal cancer cells, as well as a microcontact printing based proof-of-concept

and comparisons between 2D and 3D settings.

Chapter IX – Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic Platform, Methods of

Production and Uses Thereof: This chapter describes the new methodology to produce enzymatically-

crosslinked silk fibroin hydrogel microfluidic platform, which takes advantage of the tyrosine groups

present in the structure of silk fibroin that are known to form a covalently-crosslinked hydrogel in

amorphous and transparent state. The hydrogel based microfluidic platform was extensively

characterized in order to demonstrate its application in several TE areas, but in particular in cancer

research applications.

Section 4 – Concluding remarks

xli

Chapter X – General Conclusions and Future Perspectives: The final section of the thesis

presents the general conclusions and implications, current limitations and potential of the work

described for application in microfluidic cancer models. The main conclusions of each chapter as well

as future perspectives of the advanced strategies developed in this thesis are provided.

xlii

“Once you stop learning, you start dying”

―Albert Einstein (1879-1955)

1

SECTION 1

GENERAL INTRODUCTION

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

2

CHAPTER I

EVALUATING

BIOMATERIALS AND

MICROFLUIDICS BASED 3D

TUMOR MODELS

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

3

Chapter I

Chapter I - Evaluating Biomaterials and Microfluidics based 3D

Tumor Models1

I-1. ABSTRACT

Cancer is a major cause of morbidity and mortality worldwide, with a disease burden estimated to

increase in the coming decades. Disease heterogeneity, limited information on cancer biology and

disease mechanisms are aspects that 2D cell cultures fail to address. We review the current “state-of-

the-art” in 3D Tissue Engineering (TE) models developed for and used in cancer research. Scaffold-

based TE models and microfluidics are assessed for their potential to fill the gap between 2D models

and clinical application. Recent advances in combining the principles of 3D TE models and microfluidics

are discussed, with a special focus on biomaterials and the most promising chip-based 3D models.

Keywords: Cancer, Biomaterials, Microfluidics, 3D.

1This chapter is based on the following publication:

M.R. Carvalho, D. Lima, R. L. Reis, V. M. Correlo, J. M. Oliveira. “Evaluating biomaterials and microfluidics based 3D tumor models”.

Trends in Biotechnology, 2015, 33:667-678. doi: 10.1016/j.tibtech.2015.09.009.

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

4

I-2. THE IMPORTANCE OF 3D IN VITRO TISSUE MODELS FOR ADVANCED CANCER RESEARCH

Conventional approaches used in cancer research involve culturing of tumor cells in 2D surfaces

and the use of animal models, which poorly correlate with human disease states. 2D cell cultures

oversimplify the biological context of a tumor, which is influenced by intrinsic molecular features and

external cues from its surrounding microenvironment [1]. Unlike cancer cells grown in 2D, those grown

in 3D adopt a rounded shape, forming clusters that are suggestive of tumors in vivo [2, 3]. Cancer cells

grown in 2D versus 3D also exhibit differential gene expression profiles for key genes involved in

angiogenesis, cell migration, and invasion [4-8]. Ex vivo models or in vivo models such as animal or

patient-derived xenograft (PDX) models are also popular tools for cancer research. Such models have

advantages over cell cultures and do not suffer from the lack of 3D context, but they present their own

set of limitations (Box 1).

I-2.1. BOX 1 - Advantages and disadvantages of ex vivo models, animal models, patient derived

xenograft (PDX) models

Ex vivo models: Ex vivo tumor culture is performed using a thin slice of tumor tissue collected from

human or animal sources and cultured on porous substrates or embedded in ECM like matrices [9].

These models generally preserve the native complex and differentiated 3D cell-matrix architecture, cell

phenotype, complex architecture, logically providing a more accurate mimic of cell behavior. On the

other hand, the main drawback of this type of model may be the absence of the absence of mechanical

forces, and the lack of mechanical forces such as shear stress, perfusion and surrounding tissue may

result in changes in the structure and cell behavior when comparing to original in vivo

microenvironment. Another drawback is the need of harvesting tissue from human or animal subjects.

Animal models: Mouse models have proven essential in cancer research. These models yield

better prediction of drug behavior and efficacy in human than 2D conventional culture. They are used to

understand the genetic basis of tumor development and cancer progression. They can also be used to

test the efficacies of different anti-cancer agents as they contain their intrinsic microenvironmental

complexity. Animal models allow studies on defined mutations, including the analysis of the effects of

these mutations in many genetic backgrounds. Growing demand from the public opinion to reduce the

use of animals as experimental subjects [10-12], other limitations involve the inability to mimic human

specific features regarding tumors, autoimmune conditions, stem cell differentiation and ultimately,

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

5

their responses to therapeutic drugs because physiology, metabolism, tumor cell interactions with the

innate immune system, proliferation, metastasis and cells themselves in animals are different from

humans. [13, 14].

PDX models: Models where surgically resected primary tumor samples are engrafted directly from

patients onto immunodeficient mice. These allow preserving longer the molecular, genetic and

histological heterogeneity of their parental tumors [15]. PDX models offer a powerful tool for cancer

research and a route toward personalized medicine for cancer patients. These also enable the discovery

of biomarkers predicting drug sensitivity and resistance and possibly monitoring the initiation and

progression of metastasis as well as the fate of circulating tumor cells using in vivo flow cytometry of

implanted humor tumors [16].

To address the limitations of conventional approaches, the 3D microenvironment of tumors must

be taken into account to improve the physiological relevance of in vitro models [17, 18]. The integration

of tissue engineering (TE) strategies and microfluidic technologies has recently sparked a breakthrough

into the design of in vitro microfluidic culture models that better adapt to morphological changes in

tissue structure and function over time, providing a level of precision control that could not be achieved

before [19]. We review the current “state-of-the-art” on 3D TE models that have been developed and

used in cancer research. The relevance of 3D cell models in cancer studies is critically assessed, and

the main advantages and limitations discussed, with special emphasis on the biomaterials point of view.

New approaches that integrate bioreactors and microfluidic technology are also discussed, along with

the potential impact of 3D TE models in the cancer drug discovery process.

I-3. CLASSICAL 3D TE MODELS FOR CANCER RESEARCH

Classical 3D culture systems can be broadly subdivided as scaffold-free or scaffold-based methods

[20]. Scaffold-free 3D cancer models are best exemplified by tumor spheroids (Box 2), but we will focus

here on scaffold-based methods because they offer more opportunities for combination with other

technologies. Scaffold materials can be of synthetic or natural origin [21]. Synthetic materials typically

display better mechanical properties than natural ones (Table I-1), but we focus our discussion on

scaffolds made from naturally derived materials due to their greater physiological relevance.

Biomaterials are being broadly used for their remarkable similarities to the extracellular matrix, typically

good biological features such as, biocompatibility, biodegradability and bioavailability, as well as

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

6

inherent cellular interaction. Also, natural polymers can be engineered and their properties tuned to

obtain desirable mechanical and physical characteristics [22].

I-3.1. Box 2 - In vitro 3D multicellular tumor spheroids

Spheroids form due to the ability of cells in suspension to self-assemble when cell-cell interactions

are stronger than those between cells and substrate [23]. Their small size and the absence of

vascularization create limitations to nutrients and oxygen diffusion, requiring cells of its core to adopt a

glycolytic metabolic activity which causes an increasing pH gradient towards the outside [24]. This

mimics what happens with actual tumors, which have regions of hypoxia and of acidity similarly

structured as a function of the distance between the core and the tumor blood vessels [25-27].

Spheroids can be fabricated using several methods: i) hanging drop method, in which cells can

aggregate freely at the bottom of an inverted cell suspension drop; ii) spinner flasks method, which

provides constant agitation, allowing spontaneous cell aggregation; iii) static liquid overlay technique,

which enables spheroid formation by preventing cell adhesion to the growth substrate; iv) centrifugation;

and v) growing cells on non-adherent micropatterned surfaces in microfabricated devices to favor their

aggregation [28]. Given their tumor-like features, in vitro multicellular spheroids have been particularly

useful for studying the efficacy of novel chemotherapeutic agents or drug delivery systems. Significant

differences in drug responses have been observed for numerous cancer types in spheroid culture, but

increased chemical resistance appears to depend upon the type of cancer cells and the specific

treatment under study [29-31]. The emerging resistance to chemotherapies expands the potential

application for spheroid cancer models. Breast cancer tumor recurrence, a major cause of death, has

recently been attributed to a small population of cells with stem-like characteristics, able to self-renew

and promote tumor progression [32, 33]. A 3D spheroid tumor model with stem-like properties was

used to discover a new inhibitor of spheroid formation [34]. This stem-like enriched spheroid formation

technology could also be applied to drug discovery in other cancer types [35]. Despite the advantages

of using spheroids in cancer research, they are still not routinely incorporated into drug discovery, most

likely due to technical hurdles. For example, automated analysis systems are not compatible with loose

spheroids, which can clog pipettes and tubing. Tethered spheroids may overcome this hurdle [36].

Table I-1 - Naturally Derived Matrices for 3D Tumor Engineering.

Material Characteristics Cancer research References

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

7

applications

Matrigel® (natural)

3D hydrogel microenvironment for tumor growth and angiogenesis studies; Cytocompatibility, cell

adhesion sites, tunable physical properties.

Human colon adenocarcinoma, colorectal cancer,

prostate cancer, breast cancer;

[37, 38]

Collagen I (natural) Multiple crosslinking methods;

Biocompatibility, biodegradability, angiogenesis potential.

Human breast carcinoma, Human hepatocellular liver

carcinoma;

[31, 39-41]

Fibrinogen (natural)

Blood clotting, cellular and matrix interactions, neoplasia; Architecture

mimics native ECM, 3D microenvironment for cancer growth.

Melanoma, ovarian cancer cells, liver

carcinoma; [42, 43]

Hyaluronan (natural)

Glycosaminoglycan found in extracellular tissue in many parts of the body; Major component of the native brain extracellular matrix,

studying tumor migration processes.

Prostate cancer, Glioma tumor;

[44, 45]

Chitosan (natural) Analysis of interaction of prostate cancer tumor cells with immune

cells, formation of tumor spheroids.

Prostate cancer, glioma tumor;

[46-48]

Alginate (natural)

Properties for cell transplantation, drug delivery and tissue engineering; Suitable for hydrogel microspheres, promotes the conversion of cultured cancer cells to a more malignant in vivo-like phenotype; Non-adhesive to

cells.

Oral squamous cell carcinoma, Human

hepatocellular carcinoma;

[48-51]

Fibroblast-derived matrices (natural)

Distinctly different cell morphology, aggregation pattern, proliferation

profile and invasive potential, although these matrices do not fully

represent the composition and structure of the tumor

microenvironment.

Human colorectal carcinoma, Human

pancreatic carcinoma; [52, 53]

Silk fibroin (natural)

Unique mechanical properties, good biocompatibility, well-controlled

degradability and versatile processability.

Human breast adenocarcinoma;

[54]

Agarose (natural) Amenable mechanical and biological

properties; More stable than traditional natural hydrogels.

Osteosarcoma and breast adenocarcinoma;

[55]

Polyethylene glycol (PEG) (synthetic)

Biocompatibility, high water content, and multi-tunable properties;

Specific biological functionalities can be covalently incorporated.

Human epithelial ovarian cancer, Human

pancreatic ductal adenocarcinoma;

[56-58]

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

8

Poly(lactic-co-glycolic) acid (PLGA) (synthetic)

Highly porous scaffolds; Convenient to handle and amenable to large-

scale use.

Oral squamous, cell carcinoma;

[59, 60]

PCL (synthetic)

Biologically inert synthetic polymer, high porosity, a large surface area-to-volume ratio for cellular attachment,

tunable fiber diameter, low cost.

Ewing sarcoma; [61]

Synthetic Peptides (synthetic)

Controlled amino acid composition for easy incorporation of specific

biological relevant ligands; Adequate physiological properties.

Human ovarian carcinoma, Human breast carcinoma;

[62-64]

I-3.2. In vitro 3D scaffold-based TE tumor models

Scaffold-based models have the advantage of allowing the study of tumor interactions with the

microenvironment, in particular, phenomena such as tumor migration and invasion. Another advantage

is the possibility to functionalize the scaffold materials to obtain desired physicochemical and biological

characteristics. For example it is possible to incorporate bioactive molecules that promote cell-adhesion

or matrix metalloproteinase (MMP) substrates that render the materials susceptible to degradation by

cell-secreted proteases, thus mimicking the naturally occurring interactions of cells with ECM and its

consequent remodeling [57].

Great care and attention are required when choosing the biomaterial for culturing cancer cells, in

order to better emulate the physiology of their original ECM, since this facet alone is able to influence

tissue organization [19, 65].

Models using Matrigel® as reconstituted basement membrane [66, 67] can mimic the

pathophysiological context of cancer and have enabled advances in 3D tissue engineering. Development

of Matrigel® grew from pioneering work on the isolation and purification of proteins such laminin and

type IV procollagen and discovery of the biological activity of the reconstituted basement membrane

[68, 69]. Matrigel®, has been to mimic breast cancer progression [70] and help gain understanding of

how tissue organization itself influences the development of a malignant phenotype [67]. Despite being

the most commonly used biomaterial in cancer research, Matrigel® has some disadvantages: As it is a

biological animal derived product, it lacks human motifs (DNA with biological function); possible growth

factor contamination; possible variation of endotoxin levels and stiffness between lots and the presence

of undefined substances makes comparison between studies more challenging [71, 72].

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

9

Collagen I is a frequently used substrate for cell culture and tissue engineering applications

because it contains the tripeptide RGD (Arg-Gly-Asp), a short amino acid sequence that preferentially

binds to receptors on cell surfaces. Scaffolds made from Collagen I can be synthetically modified to

provide a wider range of physicochemical properties. For example, collagen stiffness can be adjusted

through covalent crosslinking by non-enzymatic glycation. One of the first organotypic models consisted

of isolated human fetal colonic epithelial cells seeded on a collagen type I matrix with embedded colonic

fibroblasts [73]. Collagen I was also used to produce a biologically relevant 3D tumor model that

supported unconfined cellular proliferation and exhibited necrosis beyond a depth of ≈150 m and also

had angiogenic potential [39]. The biocompatibility and 3D architecture of collagen I hydrogels are

suitable for reproducing the microenvironmental conditions of a solid tumor.

Fang et al. [49] chose alginate to recapitulate autocrine or paracrine functions in cancer. A 3D

model to study paracrine function was currently lacking, being of great relevance since paracrine

function is a major mechanism of cell-cell communication within tissue microenvironment in normal

development and disease. The system studies paracrine influence between prostate cancer cell and

normal prostate stromal cells in which direct interaction between epithelial and stromal cells is not

allowed. Alginate was the chosen biomaterial because of its hydrophilic nature allowing a high cell

loading functional in culture. Alginate hydrogels possess the adequate characteristics for cell

transplantation, drug delivery and tissue engineering as its production and cell encapsulation can be

achieved under mild conditions.

Fibrin (purified from fibrinogen) hydrogels are also used as a biomaterial in cancer research,

supporting growth of cancer cells into colonies that resembled embryonic stem cell colonies [42]. Fibrin

is present in connective tissue stroma in human malignant tumors, and fibrin and fibrinogen can

increase the survival and metastatic potential of circulating tumor cells [42]. In vitro scaffold-based

cancer models have been used in drug discovery, to understand mechanisms of action, find novel

targets, or to address drug efficacy, toxicity and resistance events. For instance, the efficacy of three

anti-cancer drugs (e.g., camptothecin, docetaxel and rapamycin) in the treatment of bone metastatic

prostate cancer has been assessed using Hyaluronic Acid (HA) derived hydrogels to grow the LNCaP

(lymph node carcinoma of the prostate) [74]. The difference in efficacy between the three drugs may

reflect their different mechanisms of action and chemical properties, emphasizing the importance of a

proper microenvironment in anti-cancer drug efficacy assessments. HA allowed cells residing in the

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

10

hydrogel matrix to form distinct clustered structures which grew and merged, reminiscent of real tumors

[74].

Other biocompatible hydrogels, such as those made of Gellan Gum or Silk [75, 76] can potentially

be used in the development of complex microtissues for cancer research. These two naturally derived

hydrogels attract a special interest since they may provide important chemical cues to the cells due to

their resemblance to the natural extracellular matrix and their ability to easily achieve a 3D model

architecture. Their biodegradability and biocompatibility have been extensively validated in vitro and in

vivo [77]. Moreover, they show tunable mechanical properties ideal for cell encapsulation. Cancer

models can therefore be created by including cancer cells, stromal fibroblasts, macrophages or growth

factors.

I-3.2.1. Novel applications for scaffolds

Scaffold based TE strategies were also applied to cancer immunotherapy to develop an in vitro 3D

scaffold model for examining the interaction of tumor-associated fibroblasts (TAF) with breast tumor

cells and breast specific, neu antigen (p98) reactive T cells [78]. Breast cancer cells seeded on 3D

chitosan-alginate (CA) scaffolds exhibited productive growth and formed distinct tumor spheroids.

Antigen specific p98 T cells, but not naïve T cells, bound better to tumor cells on scaffolds. The p98 T

cells induced potent tumor cell killing but T helper cell cytokine function was impaired in the presence

of TAF co-seeding on scaffolds. From a biomaterials perspective, CA scaffolds are important because

they are biocompatible and non-immunogenic, they have the proxy structure of glycosaminoglycans

(GAGs) (major component of the native (ECM), and they are FDA approved for numerous biomedical

applications. These kinds of strategies can bridge the gap between in vitro and pre-clinical testing of

novel immunotherapies by enabling researchers to probe individual cell types and factors in a more

physiologically relevant tumor-like microenvironment.

I-4. CHIP BASED 3D TE MODELS IN CANCER RESEARCH

TE based models still face difficult challenges, such as the requirement of functional vasculature

networks to regulate the transport of nutrients and oxygen and the need to control metabolic or

mechanical functions of the encapsulated cells within biocompatible scaffolds [79]. Combining TE

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

11

principles with microfluidic technologies has the potential to fulfill such gaps, since microfluidics allows

the fabrication of three-dimensional architectures with controlled spatial relationships between cells, the

presence of flow-induced signaling and transduction, and the capacity to introduce the chemical

gradients necessary to reproduce the architecture of the in vivo microenvironment [80-82].

The combination of TE principles and microfluidic technologies can take the form of so-called

“biochips” for 3D cell culture that better mimic the physiological environment and interactions observed

in vivo [83, 84].

In vitro multi-tissue 3D tumor models on a chip can make it possible to obtain quantitative

measurements on circulating tumor cells, extravasation and micrometastasis (Figure I-1). For example,

a microfluidic platform was built to emulate the dynamic physiology of the bone marrow

microenvironment, allowing realistic interaction of bone marrow cells and osteoblasts, so as to

investigate multiple myeloma [85].

In combination with biomaterials-based approaches, several chip-based models (Figure I-2) have

provided invaluable knowledge and previously unmeasurable or unobservable data (Figure I-3) about

cancer related processes such as angiogenesis and metastasis, and have proved instrumental in drug

discovery.

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

12

Figure I-1 - Chip-Based 3D Models in Cancer Research. Scheme of an example of a chip-based 3D model mimicking the

native extracellular matrix (ECM) tumor microenvironment. The entry of tumor cells into the blood stream

and the micrometastasis process in a variety of tissues, or in the presence of other relevant cell types, can

be studied. The chip microwells could contain a 3D engineered hydrogel representing an ECM-like matrix,

such as Matrigel®, in direct contact with microchannels mimicking blood vessels for intravasation and/or

extravasation studies. The small blue arrows represent the migration of (cancer) cells from the microchannel

and into the 3D tissue-engineered ECM-like matrix and vice-versa. (A) Cellix VenaT4chip; (B) The microwell

filled with Matrigel® (3D); (C) Microscopy image of the channel–microwell interface; (D) Representation of

cancer cells migrating from the microchannel to the 3D TE ECM-like matrix in the presence of osteoblasts;

(E) fluorescence microscopy image of HeLA cells adhered to the channel migrating into the ECM-like matrix

(Matrige®l) M. R. Carvalho, unpublished data, 2015).

I-4.1.1. Angiogenesis

Angiogenesis is a prerequisite for tumor growth, invasion, progression and metastasis, and thus

crucial to include in cancer models. Organ-on chip models that integrate vasculature have the potential

to transform in vitro approaches for the study of cancer [86, 87], offering the possibilities for spatially

resolved delivery and extraction of solvents and solutes to control the biochemistry of tumor’s

microenvironment, growth of appropriate endothelium, delivery of circulating cells, as well as the

possibility to control tension and shear stress during angiogenesis, tumor growth and drug delivery [88].

Physiological vascularized tumor conditions can be achieved by including mixed cell populations in the

device: tumor cells, stroma cells, endothelial cells that line the vessels and also immune cells. Such an

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

13

approach would enable analysis of any circulating molecular and cellular components that may promote

tumor angiogenesis [89]. The biomaterials that form the microfluidic scaffold would have to recapitulate

matrix stiffness and withstand interstitial pressure, as well as convey mechanical cues that modulate

cell signaling via mechanoreceptor signal transduction.

Microfluidic vascular models can be divided in two categories [90]: Microfabricated molds that

confine biological hydrogels between parallel microfluidic channels [91], or bona fide vascular

structures fully embedded within 3D ECM [92]. In an example of the microfabricated mold approach, a

device was developed with a central microchannel embedded within a collagen hydrogel, which allowed

tumor-relevant hydrodynamic stresses to be introduced and quantified using microparticle image

velocimetry (µ-PIV) [93].

As an example of the second approach, a 3D microphysiological system is being developed. The

model uses induced pluripotent stem cell technology (development of vessel networks derived from

human iPSC-derived ECs in a cardiac-derived ECM to simulate the microcirculation), the cardiac

muscle, and the solid tumor into a single integrated microphysiological system [94]. As a 3D matrix, a

co-culture of endothelial colony-forming-derived endothelial cells and normal human lung fibroblasts are

mixed with fibrin matrix.

Figure I-2 - Microfluidic devices used in the study of angiogenesis (A) and metastasis (B). A1) Passive pumping-based

microfluidic angiogenesis assay with 3D cylindrical lumens. (A) Illustration of a triple channel design with

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

14

connecting microchannels. (b–d) Microchannel systems can be (b) single, (c) double, or (d) triple channel

designs, and are arrayable. Reprinted with permission from [95]. Copyright, 2012 Elsevier. a2) Perfusable

three-dimensional microvessels are generated using an optically clear polydimethsiloxane microfluidic-based

platform. Reprinted with permission from [94]. Copyright, 2013 BioMed Central Ltd. b) Microfluidic

vasculature enables region-specific activation of endothelium under physiological flow conditions [96].

I-4.1.2. Metastasis

Metastasis is one of the most complex processes in cancer and likely one of the most difficult

study and mimic using in vitro models [97, 98]. Therefore it is crucial to understand the molecular and

cellular phenomena involved in the metastatic cascade [99-101]. Invasion of cancer cells through the

basal membrane into a blood or lymphatic vessel (intravasation) followed by entrance in other

tissue/organs (extravasation) are critical steps [102, 103]. Although significant progress has been

made in visualizing tumor-cell motility in vivo, the underlying mechanism of cancer cell intravasation is

largely unknown. Few studies have addressed the development of 3D models for metastasis studies

(Figure I-3). However, an in vitro 3D microfluidic model of the tumor-vascular interface was designed to

integrate live imaging, precise control of microenvironmental factors and endothelial barrier

measurement [102]. The chip consists of two independent channels in which tumor and endothelial

cells are seeded, interconnected via a 3D ECM hydrogel made of collagen type I. This work sheds light

on the influence of macrophage-secreted factors in intravasation processes. Thus, the formation of an

endothelial monolayer on a 3D collagen type I hydrogel mimicking ECM enabled the precise

quantification and control of critical microenvironmental factors.

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

15

Figure I-3 - Outputs of angiogenesis (A) and metastasis (B) studies in TE designs. A1) Cultured breast or prostate epithelial

carcinoma cells (MCF-7; LNCaP) with HUVECs and mesenchymal stromal cells (MSCs) within matrices

fabricated from synthetic starPEG and maleimide-functionalized heparin to study 3D tumor angiogenesis

microenvironments after 14 days: Extended focus confocal images displaying Phalloidin (red), Hoechst (blue)

and CD31 (green) showing HUVEC and MSC to cancer cell interactions for each tumor cell type. Reprinted

with permission from [104]. Copyright, 2015 Elsevier. A2) 3D endothelial-lined lumens (ELL) obtained by

seeding Human umbilical vein endothelial cells (HUVECs) in a microfluidic chip. Reprinted with permission

from [95]. Copyright, 2012 Elsevier. B1) Microfluidic tumor-vascular interface model: Endothelial channel

(green), tumor channel (green) and 3D ECM (grey) between the two channels. White arrow shows

Fibrosarcoma cells (HT1080) invading in 3D toward the endothelium. [102]. B2) Migration of HUVEC and

Hela cells during co-culture on-chip. Reprinted with permission from [105]. Copyright, 2012 American

Chemical Society.

I-4.1.3. Drug Discovery

Developing in vitro 3D chip-based tumor models will not only aid in investigating angiogenesis and

metastasis, but can help provide a realistic preclinical assessment of anti-cancer drug efficacy and

resistance. For example, a microfluidic chip-based, 3D co-culture drug sensitivity test platform was

developed, in which a mono-lung cancer cell line, a mixture of lung cancer and stromal cell lines, and

cells from fresh lung cancer tissues were and treated with anti-cancer drugs [106]. A gradient

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

16

concentration generator inside the chips allowed reconstructing tumor microenvironments in vitro with

continuous nutrient and oxygen supplementation. Cell culture medium was introduced into the

microchannels to generate a set of gradient concentrations for each drug or combination of drugs.

Moreover, these drug sensitivity tests were carried out on the fresh tissues of cancer which allowed

screening single or combination anti-cancer drug schemes efficiently and accurately for real patients.

In addition to the chips independently developed by researchers in academic laboratories,

companies are emerging that produce microfluidic-based 3D cell culture devices that are approved for

commercialization. For example, MIMETAS produces OrganoPlates™ which are microfluidics‐based

culture plates that enable culturing and screening of a large range of organ and tissue models. These

platforms enable precise deposition of cells in a 3D culture matrix, and their patented liquid handling

technology allows for better readout and quantification than conventional culture systems. The cells are

contained in a gel that allows 3D tissue configurations and cell-cell interactions. A continuous perfusion

of media through the plate mimics blood flow and exchange of nutrients, oxygen and metabolites.

Other companies in the market are also following this new trend. Cellix’s VenaT4™ chip is suitable

for ECM-matrix, collagen gels, hydrogels, Matrigel® or similar aqueous biomaterials and ideal for

invasion assays in an in vivo like settings. Such companies will revolutionize the field by working closely

with researchers and clinicians to develop a customized in vivo based accurate biomimicry.

I-5. CONCLUDING REMARKS AND FUTURE PERSPECTIVES

Bioengineered 3D microsystem technologies are relatively new and still require great effort to

validate and characterize their properties, and suitability for practical biomedical applications. Despite

the promise of microfluidic chip-based 3D cell culture systems for cancer research, there are some

disadvantages compared to conventional techniques, and there are challenges that need to be

addressed. Moving in vitro culture from macroscopic culture to PDMS based devices can come with

unforeseen challenges. Changes in device material, surface coating, cell number per unit surface area

or per unit media volume may all affect the outcome of otherwise standard protocols, and in this sense,

surface materials and treatments deserve special attention. Although these systems present numerous

advantages in terms of their ability to mimic what happens in vivo and to visualize cell growth, cell

migration and cell-cell interactions, the small volumes used as well as the low numbers of cells within

the microfluidic devices make conventional biochemical assays more challenging due to detection

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

17

limits, making it difficult to generate statistically significant amounts of data. Advancements in

technologies that are sensitive enough to detect all-important genetic and transcriptomic changes are

required for the discovery of novel biomarkers and critical events in cancer development. Regardless,

their potential predicting clinical responses could have great effects on the way drug discovery and

bioequivalence studies, pathogenesis of relevant diseases such as cancer can be investigated.

Further development of chip-based 3D cell culture in cancer research will be largely dependent on

the improvement of biomaterials that emulate the ECM, and the capacity to scale-up these complex

technologies. Integrating TE approaches and microfluidics into easy-to-use, scalable, reproducible and

cost-effective systems will be the key to their success and future translation to the market. A 3D multi-

tissue in vitro tumor model on a chip could contribute to accelerating the time-to-market for anti-cancer

drugs, along with a well-defined regulatory and development strategy. 3D in vitro models could

demonstrate whether different formulations of the same drug are bioequivalent. A straightforward

bioequivalence trial comparing relevant pharmacokinetic parameters of both formulations would be

instrumental in gaining FDA approval.

There is now a huge demand for new strategies and more suitable biomaterials to interface with

microfluidic chips for cell biological studies. Clearly, there is still a long road ahead for this to become

reality, but we believe this is the right direction to pursue in the search for novel and more efficacious

treatments for cancer.

I-6. REFERENCES

1. Albini, A. and M.B. Sporn, The tumour microenvironment as a target for chemoprevention. Nat Rev Cancer, 2007. 7(2): p. 139-47. 2. Fang, Q.B., et al., Three Overlapping Balloon-expandable Stents Applied in Coeliac Artery Dissection: Case Report and Literature Review. West Indian Med J, 2014. 63(7): p. 791-2. 3. Gurski, L.A., et al., 3D matrices for anti-cancer drug testing and development. Oncology Issues, 2010. 25: p. 20-25. 4. Cheema, U., et al., Spatially defined oxygen gradients and vascular endothelial growth factor expression in an engineered 3D cell model. Cell Mol Life Sci, 2008. 65(1): p. 177-86. 5. Kenny, P.A., et al., The morphologies of breast cancer cell lines in three-dimensional assays correlate with their profiles of gene expression. Mol Oncol, 2007. 1(1): p. 84-96. 6. Kurisu, S., et al., Rac-WAVE2 signaling is involved in the invasive and metastatic phenotypes of murine melanoma cells. Oncogene, 2005. 24(8): p. 1309-19. 7. Ridky, T.W., et al., Invasive three-dimensional organotypic neoplasia from multiple normal human epithelia. Nat Med, 2010. 16(12): p. 1450-5.

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

18

8. Wozniak, M.A., et al., Focal adhesion regulation of cell behavior. Biochim Biophys Acta, 2004. 1692(2-3): p. 103-19. 9. Goodman, T.T., C.P. Ng, and S.H. Pun, 3-D tissue culture systems for the evaluation and optimization of nanoparticle-based drug carriers. Bioconjug Chem, 2008. 19(10): p. 1951-9. 10. Lee, M.Y., et al., Three-dimensional cellular microarray for high-throughput toxicology assays. Proc Natl Acad Sci U S A, 2008. 105(1): p. 59-63. 11. Meli, L., et al., Influence of a three-dimensional, microarray environment on human cell culture in drug screening systems. Biomaterials, 2012. 33(35): p. 9087-96. 12. Mehling, A., et al., Non-animal test methods for predicting skin sensitization potentials. Arch Toxicol, 2012. 86(8): p. 1273-95. 13. Yamada, K.M. and E. Cukierman, Modeling tissue morphogenesis and cancer in 3D. Cell, 2007. 130(4): p. 601-10. 14. Rangarajan, A. and R.A. Weinberg, Opinion: Comparative biology of mouse versus human cells: modelling human cancer in mice. Nat Rev Cancer, 2003. 3(12): p. 952-9. 15. Hao, C., et al., Gene mutations in primary tumors and corresponding patient-derived xenografts derived from non-small cell lung cancer. Cancer Letters, 2015. 357(1): p. 179-185. 16. Hwu, D., et al., Assessment of the role of circulating breast cancer cells in tumor formation and metastatic potential using in vivo flow cytometry. J Biomed Opt, 2011. 16(4): p. 040501. 17. Jason, E.E., et al., Three-Dimensional Lung Tumor Microenvironment Modulates Therapeutic Compound Responsiveness In Vitro – Implication for Drug Development. PLoS One, 2014. 9(3). 18. Hsiao, A.Y., et al., Microfluidic system for formation of PC-3 prostate cancer co-culture spheroids. Biomaterials, 2009. 30(16): p. 3020-7. 19. Elliott, N.T. and F. Yuan, A review of three-dimensional in vitro tissue models for drug discovery and transport studies. J Pharm Sci, 2011. 100(1): p. 59-74. 20. Rimann, M. and U. Graf-Hausner, Synthetic 3D multicellular systems for drug development. Curr Opin Biotechnol, 2012. 23(5): p. 803-9. 21. Peck, Y. and D.A. Wang, Three-dimensionally engineered biomimetic tissue models for in vitro drug evaluation: delivery, efficacy and toxicity. Expert Opin Drug Deliv, 2013. 10(3): p. 369-83. 22. Mano, J.F., Designing biomaterials for tissue engineering based on the deconstruction of the native cellular environment. Materials Letters, 2015. 141: p. 198-202. 23. Liang, Y., et al., A cell-instructive hydrogel to regulate malignancy of 3D tumor spheroids with matrix rigidity. Biomaterials, 2011. 32(35): p. 9308-9315. 24. D'Souza, S., et al., Gfi1 expressed in bone marrow stromal cells is a novel osteoblast suppressor in patients with multiple myeloma bone disease. Blood, 2011. 118(26): p. 6871-80. 25. Gatenby, R.A., et al., Cellular adaptations to hypoxia and acidosis during somatic evolution of breast cancer. Br J Cancer, 2007. 97(5): p. 646-53. 26. Kazmi, N., et al., Avascular tumour growth dynamics and the constraints of protein binding for drug transportation. J Theor Biol, 2012. 313: p. 142-52. 27. Tredan, O., et al., Drug resistance and the solid tumor microenvironment. J Natl Cancer Inst, 2007. 99(19): p. 1441-54. 28. Fennema, E., et al., Spheroid culture as a tool for creating 3D complex tissues. Trends Biotechnol, 2013. 31(2): p. 108-15. 29. Tung, Y.C., et al., High-throughput 3D spheroid culture and drug testing using a 384 hanging drop array. Analyst, 2011. 136(3): p. 473-8. 30. Wen, Z., et al., A spheroid-based 3-D culture model for pancreatic cancer drug testing, using the acid phosphatase assay. Braz J Med Biol Res, 2013. 46(7): p. 634-42.

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

19

31. Orlandi, P., et al., Idarubicin and idarubicinol effects on breast cancer multicellular spheroids. J Chemother, 2005. 17(6): p. 663-7. 32. Clarke, M.F. and M. Fuller, Stem cells and cancer: two faces of eve. Cell, 2006. 124(6): p. 1111-5. 33. Reya, T., et al., Stem cells, cancer, and cancer stem cells. Nature, 2001. 414(6859): p. 105-11. 34. Wang, Y.C., et al., Drug screening identifies niclosamide as an inhibitor of breast cancer stem-like cells. PLoS One, 2013. 8(9): p. e74538. 35. Jung, P., et al., Isolation and in vitro expansion of human colonic stem cells. Nat Med, 2011. 17(10): p. 1225-7. 36. Shim, M.S., et al., A bioreducible polymer for efficient delivery of Fas-silencing siRNA into stem cell spheroids and enhanced therapeutic angiogenesis. Angew Chem Int Ed Engl, 2012. 51(47): p. 11899-903. 37. Yeung, T.M., et al., Cancer stem cells from colorectal cancer-derived cell lines. Proc Natl Acad Sci U S A, 2010. 107(8): p. 3722-7. 38. Benton, G., et al., Multiple uses of basement membrane-like matrix (BME/Matrigel) in vitro and in vivo with cancer cells. Int J Cancer, 2011. 128(8): p. 1751-7. 39. Szot, C.S., et al., 3D in vitro bioengineered tumors based on collagen I hydrogels. Biomaterials, 2011. 32(31): p. 7905-12. 40. Chen, L., et al., The enhancement of cancer stem cell properties of MCF-7 cells in 3D collagen scaffolds for modeling of cancer and anti-cancer drugs. Biomaterials, 2012. 33(5): p. 1437-44. 41. Yip, D. and C.H. Cho, A multicellular 3D heterospheroid model of liver tumor and stromal cells in collagen gel for anti-cancer drug testing. Biochem Biophys Res Commun, 2013. 433(3): p. 327-32. 42. Liu, J., et al., Soft fibrin gels promote selection and growth of tumorigenic cells. Nat Mater, 2012. 11(8): p. 734-41. 43. Mosesson, M.W., Fibrinogen and fibrin structure and functions. J Thromb Haemost, 2005. 3(8): p. 1894-904. 44. Rao, S.S., et al., Glioblastoma behaviors in three-dimensional collagen-hyaluronan composite hydrogels. ACS Appl Mater Interfaces, 2013. 5(19): p. 9276-84. 45. Xu, X., et al., Recreating the tumor microenvironment in a bilayer, hyaluronic acid hydrogel construct for the growth of prostate cancer spheroids. Biomaterials, 2012. 33(35): p. 9049-60. 46. Berger, J., et al., Structure and interactions in covalently and ionically crosslinked chitosan hydrogels for biomedical applications. Eur J Pharm Biopharm, 2004. 57(1): p. 19-34. 47. Florczyk, S.J., et al., 3D porous chitosan-alginate scaffolds: a new matrix for studying prostate cancer cell-lymphocyte interactions in vitro. Adv Healthc Mater, 2012. 1(5): p. 590-9. 48. Kievit, F.M., et al., Chitosan-alginate 3D scaffolds as a mimic of the glioma tumor microenvironment. Biomaterials, 2010. 31(22): p. 5903-10. 49. Fang, X., et al., Novel 3D co-culture model for epithelial-stromal cells interaction in prostate cancer. PLoS One, 2013. 8(9): p. e75187. 50. Fischbach, C., et al., Cancer cell angiogenic capability is regulated by 3D culture and integrin engagement. Proc Natl Acad Sci U S A, 2009. 106(2): p. 399-404. 51. Xu, X.X., et al., Encapsulated human hepatocellular carcinoma cells by alginate gel beads as an in vitro metastasis model. Exp Cell Res, 2013. 319(14): p. 2135-44. 52. Nyga, A., U. Cheema, and M. Loizidou, 3D tumour models: novel in vitro approaches to cancer studies. J Cell Commun Signal, 2011. 5(3): p. 239-48. 53. Serebriiskii, I., et al., Fibroblast-derived 3D matrix differentially regulates the growth and drug-responsiveness of human cancer cells. Matrix Biol, 2008. 27(6): p. 573-85.

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

20

54. Talukdar, S. and S.C. Kundu, A Non-Mulberry Silk Fibroin Protein Based 3D In Vitro Tumor Model for Evaluation of Anticancer Drug Activity. Advanced Functional Materials, 2012. 22(22): p. 4778-4788. 55. Xu, G., et al., In vitro ovarian cancer model based on three-dimensional agarose hydrogel. J Tissue Eng, 2014. 5: p. 2041731413520438. 56. Gill, B.J., et al., A synthetic matrix with independently tunable biochemistry and mechanical properties to study epithelial morphogenesis and EMT in a lung adenocarcinoma model. Cancer Res, 2012. 72(22): p. 6013-23. 57. Loessner, D., et al., Bioengineered 3D platform to explore cell-ECM interactions and drug resistance of epithelial ovarian cancer cells. Biomaterials, 2010. 31(32): p. 8494-506. 58. Ki, C.S., H. Shih, and C.C. Lin, Effect of 3D matrix compositions on the efficacy of EGFR inhibition in pancreatic ductal adenocarcinoma cells. Biomacromolecules, 2013. 14(9): p. 3017-26. 59. Fischbach, C., et al., Engineering tumors with 3D scaffolds. Nat Methods, 2007. 4(10): p. 855-60. 60. Ho, W.J., et al., Incorporation of multicellular spheroids into 3-D polymeric scaffolds provides an improved tumor model for screening anticancer drugs. Cancer Sci, 2010. 101(12): p. 2637-43. 61. Fong, E.L., et al., Modeling Ewing sarcoma tumors in vitro with 3D scaffolds. Proc Natl Acad Sci U S A, 2013. 110(16): p. 6500-5. 62. Huang, H., et al., Peptide hydrogelation and cell encapsulation for 3D culture of MCF-7 breast cancer cells. PLoS One, 2013. 8(3): p. e59482. 63. Xu, X., M.C. Farach-Carson, and X. Jia, Three-dimensional in vitro tumor models for cancer research and drug evaluation. Biotechnol Adv, 2014. 32(7): p. 1256-68. 64. Yang, Z. and X. Zhao, A 3D model of ovarian cancer cell lines on peptide nanofiber scaffold to explore the cell-scaffold interaction and chemotherapeutic resistance of anticancer drugs. Int J Nanomedicine, 2011. 6: p. 303-10. 65. Schmeichel, K.L. and M.J. Bissell, Modeling tissue-specific signaling and organ function in three dimensions. J Cell Sci, 2003. 116(Pt 12): p. 2377-88. 66. Weaver, V.M., et al., Reversion of the malignant phenotype of human breast cells in three-dimensional culture and in vivo by integrin blocking antibodies. J Cell Biol, 1997. 137(1): p. 231-45. 67. Li, Q., et al., p21-Activated kinase 1 coordinates aberrant cell survival and pericellular proteolysis in a three-dimensional culture model for premalignant progression of human breast cancer. Neoplasia, 2008. 10(4): p. 314-29. 68. Terranova, V.P., et al., Regulation of cell attachment and cell number by fibronectin and laminin. J Cell Physiol, 1986. 127(3): p. 473-9. 69. Kleinman, H.K., et al., Isolation and characterization of type IV procollagen, laminin, and heparan sulfate proteoglycan from the EHS sarcoma. Biochemistry, 1982. 21(24): p. 6188-93. 70. Kenny, P.A. and M.J. Bissell, Targeting TACE-dependent EGFR ligand shedding in breast cancer. J Clin Invest, 2007. 117(2): p. 337-45. 71. Wong, A.P., et al., Partitioning microfluidic channels with hydrogel to construct tunable 3-D cellular microenvironments. Biomaterials, 2008. 29(12): p. 1853-61. 72. Lee, G.Y., et al., Three-dimensional culture models of normal and malignant breast epithelial cells. Nat Methods, 2007. 4(4): p. 359-65. 73. Kalabis, J., et al., Stimulation of human colonic epithelial cells by leukemia inhibitory factor is dependent on collagen-embedded fibroblasts in organotypic culture. FASEB J, 2003. 17(9): p. 1115-7. 74. Gurski, L.A., et al., Hyaluronic acid-based hydrogels as 3D matrices for in vitro evaluation of chemotherapeutic drugs using poorly adherent prostate cancer cells. Biomaterials, 2009. 30(30): p. 6076-85.

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

21

75. Silva-Correia, J., et al., Angiogenic potential of gellan-gum-based hydrogels for application in nucleus pulposus regeneration: in vivo study. Tissue Eng Part A, 2012. 18(11-12): p. 1203-12. 76. Yan, L.P., et al., Tumor Growth Suppression Induced by Biomimetic Silk Fibroin Hydrogels. Sci Rep, 2016. 6: p. 31037. 77. Yan, L.P., et al., In vitro evaluation of the biological performance of macro/micro-porous silk fibroin and silk-nano calcium phosphate scaffolds. J Biomed Mater Res B Appl Biomater, 2015. 103(4): p. 888-98. 78. Phan-Lai, V., et al., Three-dimensional scaffolds to evaluate tumor associated fibroblast-mediated suppression of breast tumor specific T cells. Biomacromolecules, 2013. 14(5): p. 1330-7. 79. Chung, B.G., et al., Microfluidic fabrication of microengineered hydrogels and their application in tissue engineering. Lab Chip, 2012. 12(1): p. 45-59. 80. Inamdar, N.K. and J.T. Borenstein, Microfluidic cell culture models for tissue engineering. Curr Opin Biotechnol, 2011. 22(5): p. 681-9. 81. Chen, Y.A., et al., Generation of oxygen gradients in microfluidic devices for cell culture using spatially confined chemical reactions. Lab Chip, 2011. 11(21): p. 3626-33. 82. Zhou, Y. and Q. Lin, Microfluidic flow-free generation of chemical concentration gradients. Sensors and Actuators B: Chemical, 2014. 190(0): p. 334-341. 83. Choi, Y., et al., Three dimensional MEMS microfluidic perfusion system for thick brain slice cultures. Biomed Microdevices, 2007. 9(1): p. 7-13. 84. Toh, Y.C., et al., A novel 3D mammalian cell perfusion-culture system in microfluidic channels. Lab Chip, 2007. 7(3): p. 302-9. 85. Zhang, W., et al., Patient-Specific 3D Microfluidic Tissue Model for Multiple Myeloma. Tissue Eng Part C Methods, 2014. 86. Young, E.W., Cells, tissues, and organs on chips: challenges and opportunities for the cancer tumor microenvironment. Integr Biol (Camb), 2013. 5(9): p. 1096-109. 87. Chrobak, K.M., D.R. Potter, and J. Tien, Formation of perfused, functional microvascular tubes in vitro. Microvasc Res, 2006. 71(3): p. 185-96. 88. Jain, R.K., Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science, 2005. 307(5706): p. 58-62. 89. Bergers, G., et al., Benefits of targeting both pericytes and endothelial cells in the tumor vasculature with kinase inhibitors. J Clin Invest, 2003. 111(9): p. 1287-95. 90. Seo, B.R., P. DelNero, and C. Fischbach, In vitro models of tumor vessels and matrix: Engineering approaches to investigate transport limitations and drug delivery in cancer. Advanced Drug Delivery Reviews, 2014. 69–70(0): p. 205-216. 91. Nguyen, D.H., et al., Biomimetic model to reconstitute angiogenic sprouting morphogenesis in vitro. Proc Natl Acad Sci U S A, 2013. 110(17): p. 6712-7. 92. Miller, J.S., et al., Rapid casting of patterned vascular networks for perfusable engineered three-dimensional tissues. Nat Mater, 2012. 11(9): p. 768-74. 93. Buchanan, C.F., et al., Three-dimensional microfluidic collagen hydrogels for investigating flow-mediated tumor-endothelial signaling and vascular organization. Tissue Eng Part C Methods, 2014. 20(1): p. 64-75. 94. Moya, M., D. Tran, and S.C. George, An integrated in vitro model of perfused tumor and cardiac tissue. Stem Cell Res Ther, 2013. 4 Suppl 1: p. S15. 95. Bischel, L.L., et al., Tubeless microfluidic angiogenesis assay with three-dimensional endothelial-lined microvessels. Biomaterials, 2013. 34(5): p. 1471-7. 96. Song, J.W., et al., Microfluidic endothelium for studying the intravascular adhesion of metastatic breast cancer cells. PLoS One, 2009. 4(6): p. e5756.

Chapter I – Evaluating Biomaterials and Microfluidics based 3D Tumor Models

22

97. Harris, D.T., Hormonal therapy and chemotherapy of renal-cell carcinoma. Semin Oncol, 1983. 10(4): p. 422-30. 98. Shablak, A., et al., T cell-based immunotherapy of metastatic renal cell carcinoma: modest success and future perspective. Clin Cancer Res, 2009. 15(21): p. 6503-10. 99. Pouliot N, P.B., Burrows A, Investigating Metastasis Using In Vitro Platforms. In: Madame Curie Bioscience Database. 2000, Landes Bioscience: Austin (TX). 100. Valastyan, S. and R.A. Weinberg, Tumor metastasis: molecular insights and evolving paradigms. Cell, 2011. 147(2): p. 275-92. 101. Chaffer, C.L. and R.A. Weinberg, A perspective on cancer cell metastasis. Science, 2011. 331(6024): p. 1559-64. 102. Zervantonakis, I.K., et al., Three-dimensional microfluidic model for tumor cell intravasation and endothelial barrier function. Proc Natl Acad Sci U S A, 2012. 109(34): p. 13515-20. 103. Bersini, S., et al., A microfluidic 3D in vitro model for specificity of breast cancer metastasis to bone. Biomaterials, 2014. 35(8): p. 2454-61. 104. Bray, L.J., et al., Multi-parametric hydrogels support 3D in vitro bioengineered microenvironment models of tumour angiogenesis. Biomaterials, 2015. 53(0): p. 609-620. 105. Zheng, C., et al., Quantitative study of the dynamic tumor-endothelial cell interactions through an integrated microfluidic coculture system. Anal Chem, 2012. 84(4): p. 2088-93. 106. Xu, Z., et al., Application of a microfluidic chip-based 3D co-culture to test drug sensitivity for individualized treatment of lung cancer. Biomaterials, 2013. 34(16): p. 4109-17.

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

23

Chapter II

Biomaterials and

Microfluidics for Drug

Discovery and Development

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

24

Chapter II

Chapter II - Biomaterials and Microfluidics for Drug

Discovery and Development1

ABSTRACT

Microfluidic devices are one of the most promising tools to mimic in vivo like conditions, either in

normal or disease scenarios, such as tumorigenesis or pathogenesis. The combination of microfluidics

with biomaterials represents the ability to more closely mimic cells’ natural microenvironment

concerning its three-dimensional (3D) nature and continuous perfusion with nutrients and cells’

crosstalk. Due to miniaturization and increased experimental throughput, microfluidics have generated

significant interest in the drug discovery and development domain. Herein, the most recent advances in

the field of microfluidics for drug discovery are overviewed, and the role of biomaterials in 3D in vitro

models and the contribution of organ-on-a-chip technologies highlighted.

Keywords: Drug discovery; Microfluidics; Organ-on-chip.

1This chapter is based on the following publication:

M. R. Carvalho, R. Truckenmüller, R. L. Reis J. M. Oliveira, Biomaterial- and Microfluidic-Based 3D Models, Springer, Ed: J. Miguel Oliveira

and Rui L. Reis (2019).

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

25

II-1. INTRODUCTION

Many diseases, particularly acute disorders, are now treatable or manageable very effectively. The

discovery of new medications for health problems that affect a great part of the world’s population, such

as cancer, cardiovascular diseases, metabolic disorders, arthritis, depression, anxiety, gastrointestinal

disorders, pain, infectious diseases, and many others, have led to an improvement in overall health [1].

As a result, the increased quality of life may lead to increased life expectancy [1].

The drug discovery process traditionally starts in the laboratory where cells are cultured on flat

bottom cell culture flasks or Petri dishes and their response monitored upon the addition of a drug. If

successful, the experimentation progresses through the established chain of drug development: animal

and human clinical trials. Only then the product can be approved by the regulatory agencies, such as

the Food and Drug Administration (FDA; US) or the European Medicines Agency (EMEA; EU), and

commercialized [2]. However, it is well described in the literature that the vast majority of “hits” fail

between animal testing and clinical trials (around 90 %). Among the primary causes of failure, non-

clinical/clinical safety (>50 %) and efficacy (>10 %) stand out from all other factors (e.g., strategic,

commercial or operational) [3]. Finally, the drug that is successful in clinical assays takes between 12

to15 years to reach the market, and with a cost that can exceed $1 billion [4, 5].

Animal experimentation is still indispensable for evaluating the efficiency of drugs in terms of their

adsorption, distribution, metalolism, excretion, and toxicity (ADMET). However, their present certain

ethical and, in particular, clinical issues. This is because animal models, besides displaying a highly

complex environment, they are not predictive of the effects of a drugs in humans. As an examples,

studies regarding the efficiency of cancer drugs report that the average rate of successful translation

from animal models to clinical cancer trials is less than 8 % [5, 6].

In order to improve the drug development pipeline, novel cell-based high-throughput screening

(HTS) techniques have been developed. This have enabled the screening of more than a million

compounds in a couple of months [2]. However, still only about 1 out of 5.000 promising drugs

successfully makes it to the market. Even though HTS improves the screening speed and is now

considered a standard platform for drug discovery, it leaves several unresolved issues, being the

monetary costs and efficiency on top. Currently, pharmaceutical industries employ HTS primarily for

chemical optimization at the early stage of the drug development [7]. Recent estimates predict that HTS

products only contribute to 19–33 % (or less) to successful marketed drugs [8]. This may be the result

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

26

of the still limited performance of most of these conventional screening platforms due to their reliance

on flat and static culture maintenance. To tackle these issues, 3D cell-based in vitro assays have been

actively pursued, most of the times using hydrogels [9].

As a promising solution, microfluidic technologies emerged, in this case those also providing 3D

microenvironments with microvascular-like perfusion and diffusion between mimicked micro-vessels

and 3D cell culture, which is closer to what cells encounter in real tissues or organs.

Briefly, microfluidics is the manipulation of low volumes of fluids in channels with dimensions of

tens of micrometers [10]. Microfluidic chips are small platforms typically made of elastomeric materials

and comprising channel systems connected to liquid reservoirs by, for example, tubing systems in turn

linked to syringes. [11].

In combination with appropriate biological assays and high-sensitivity detection techniques,

microfluidic-based systems allow the identification and isolation of individual cells or molecules [12].

Indeed, (bio)sensors, detectors and optical components can be easily integrated on chip [13]. A great

variety of microchip fabrication techniques and materials are available for producing highly

sophisticated 2½D and 3D microstructures with integrated modules.

The high performance displayed by microfluidic-based systems holds very promising for biomedical

research and, in particular, for cancer research. This is demonstrated by the rapid market growth of

microfluidic-related products during the last years, with an emphasis on the translational research. In

particular, currently in the USA, there are three companies using microfluidic chips for in vivo drug

delivery, five microchip-based cancer cell/biomarker diagnostic companies, and about 10 companies

working on microfabricated cell-based chips for drug screening and discovery applications [14]. The

same market is growing also in Europe with very successful companies such as MIMETAS, Cellesce,

Biond Solutions and Astraveus, among others (http://ufluidix.com/circle/microfluidic-companies/)

(Assessed on 12/08/2019).

This chapter focuses on current and future applications of biomaterials in microfluidics,

microfluidics systems in drug discovery and highlights the potential of “organ-on-a-chip” technology for

drug discovery. Organ-on-a-chip are systems developed by combining microfluidic technologies and

tissue engineering to mimic physiological conditions in the body [15].

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

27

II-2. BIOMATERIALS APPLIED IN MICROFLUIDIC SYSTEMS

Biomaterials are gradually being developed as in vitro microenvironments mimicking in vivo cell

niches. They are also widely used in tissue engineering for the regeneration and modeling of a diverse

variety of both normal and diseased tissues [16]. The development of 3D culture models with tunable

biomaterials has facilitated the investigation of biological phenomena with a higher level of complexity

and physiological relevance than the traditional 2D models [17].

The combination of microfluidic technologies and biomaterials enables the reproduction of

important spatiotemporal features of complex native scenario. This facilitates the identification of new

mechanisms of cell regulation and pathogenesis, profoundly impacting on drug discovery [18].

Among all the type of 3D biomaterials, hydrogels display advanced capabilities for their integration

into microfluidic systems. Briefly, hydrogels can be derived from natural polymers or from synthetic

origin. The formers include chitosan, alginate, hyaluronic acid (HA), gellan gum (GG), agarose, collagen,

gelatin, or fibroin [19]. The latter include poly(ethylene glycol) (PEG), polyvinyl alcohol (PVA),

poly(N,Ndimethylacrylamide), and methoxy poly(ethylene glycol)-poly(e-caprolactone), among others

[20-24].

Cross-linked hydrogel networks can be engineered on a molecular basis to tune their

physicochemical properties, such as their porosity or mechanical properties such as [25]. 3D hydrogel-

based biomaterials, both natural and synthetic, are therefore able to match the physical, chemical, and

mechanical properties of the native (ECM) network, justifying their widespread use as ECM mimics in

microfluidic models [26, 27].

Various types of hydrogels have been used in microfluidic systems [28] being collagen [29, 30],

PEG [18, 31], agarose [32], fibrin [30, 33], and MatrigelTM [34, 35] some of the most commonly

employed.Toh et al. [30] used collagen to develop a microfluidic 3D hepatocyte chip (‘3D HepaTox

Chip’) for in vitro drug toxicity testing. The 3D HepaTox Chip was based on multiplexed microfluidic

channels where a 3D microenvironment was engineered in each of the eight channels to maintain the

hepatocytes’ metabolic functions [30]. Positively-charged methylated collagen and negatively-charged

HEMA-MMA-MAA terpolymer were chosen to form a localized 3D matrix due to their ability to support

hepatocyte functions.

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

28

Dinh et al. [31] combined PEG with microfluidics for precision single-cell handling. The resulting

model comprised a compartmentalized neuronal array (CNA) microfluidic circuit that allowed high-

throughput experimentation (Figure II-1) [31]. Similarly, Gao et al. [36] employed photolithography to

geometrically encapsulate human hepatoma HepG2 and human lung epithelial A549 cells in a 3D PEG

hydrogel inside a microfluidic device. The cells in the device were exposed to anticancer drugs and

apoptosis was studied (Figure II-2). Two anticancer drugs (actinomycin D and methotrexate) exhibited

distinct effects on the levels of intracellular glutathione and reactive oxygen species, indicating the

selectivity of these drugs on the disturbance of redox balance within cells [36].

To address the global challenge of vascularization, Zhang et al. [37] employed additive

manufacturing to build a blood vessel-like microfluidic structure, which was then embedded inside the

bulk material. To achieve this, alginate and chitosan hydrogels were used for direct printing of the

channels with a wall thickness below 200 μm. The printed microfluidic network was able to sustain

media perfusion, both without and with being embedded in bulk hydrogels to support cell viability,

representing an improvement towards vascularization models [37].

Regarding fibrin hydrogels, Carrion et al. [33] described a novel 3D microfluidic device as a model

system to study the molecular regulation of perivascular stem cell niches. Endothelial cells (ECs)

suspended within 3D fibrin hydrogels patterned in the device adjacent to stromal cells (either fibroblasts

or human bone marrow-derived mesenchymal stromal cells, hBM-MSCs) executed a morphogenetic

process similar to vasculogenesis, maturing into a robust capillary network with well-defined hollow

lumens. Both MSCs and fibroblasts formed pericytic associations with the ECs but promoted capillary

morphogenesis with distinct kinetics [33].

The fields of biomaterials engineering and microfluidics have advanced to impressive levels of

sophistication. The pioneering efforts reported above to further integrate biomaterials technology into

microfluidic chips tackle pertinent questions in several fields ranging from drug discovery to stem cell

research.

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

29

Figure II-1 - Biomaterials applied in microfluidic systems. 1) In situ neuron patterning. Differentiated human SH-SY5Y

neurons cultured for 5 days within the CNA devices on a PL/PLL-g-PEG array of micro-sized channels (A) and

on a continuous PL coating (B). Biomaterial patterning increases neurite guidance into the outgrowth

microchannels. Reprinted with permission from [31]. Copyright © 2018 Royal Society of Chemistry. 2 - C)

Optical image of hydrogel microstructure arrays containing two phenotypes of cells: HepG2 and A549 cells.

HepG2 cells were encapsulated inside cylindrical hydrogel microstructures (78 μm Ø × 50 μm), while A549

cells were encapsulated inside rectangular hydrogel microstructures (330 μm × 217 μm × 50 μm). D)

Corresponding fluorescence image for the cell viability expressed by live/dead assay (live cells green, dead

cells red). Reprinted with permission from [36]. Copyright © 2010 Elsevier.

ECM-derived biomaterials approaches are currently being studied for their potential to preserve the

tissue-specific biochemical composition as well as the ultrastructure of the native ECM [38]. The

development of biomaterials that promote the formation of functional tissues in clinical applications is

therefore of the utmost importance in tissue engineering and cancer research. So far, many studies

have focused on the decellularization of tissues and organs, including small intestinal submucosa, heart

valve, blood vessel, skin, nerve, tendon, ligament, bladder, amniotic membrane, heart, liver and lung

[39-41]. For example, Choi et al. [40] decellularized an ECM scaffold derived from porcine cartilage and

proved it was more efficient in maintaining chondrogenesis of rabbit mesenchymal stem cells (rMSCs)

in vitro and in the nude mouse model in vivo than the PGA scaffold used as a control [40].

Uygun et al. [39] generated a transplantable liver graft using perfusion decellularization technique,

and further introduced perfusion-seeding and culture techniques for the preparation of recellularized

liver matrix for transplantation. The recellularized graft supported liver-specific function including

albumin secretion, urea synthesis and cytochrome P450 expression at comparable levels to normal

liver in vitro [39]. Inorganic compounds, such as hydroxyapatite, have also been integrated into

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

30

microfluidic models for mimicking bone tissues. For example, Jusoh et al. [42] developed a microfluidic

platform that integrates fibrin ECM with the synthetic bone mineral hydroxyapatite nanoparticles to

provide in vivo-like microenvironments for bone vessel sprouting. Results showed that fibrin with 0.2 %

hydroxyapatite exhibited a higher number and length of sprouts [42].

II-3. MICROFLUIDICS IN DRUG DISCOVERY

Microfluidics represent an improvement to existing technologies in key areas of drug discovery.

Microfluidics involves the use of materials and techniques for controlling the movement of minute

quantities of fluids. Its ability to miniaturize assays and increase experimental throughput have

generated significant interest of the drug discovery and development domain [43].

Microfluidics can be applied to drug discovery in a variety of ways. For instance, it can be used in

(a) target identification, capture, and analysis, or (b) for developing dynamic cell culture platforms in

which cell response to a drug is assessed in a more in vivo-like environment [44]. In the following, we

briefly describe these two applications:

a) Target identification

In drug discovery, the signal transduction pathways and protein-protein interactions within cells

must be understood. Microfluidic devices can be used to address this issue by manipulating, labeling,

lysing, separating and quantifying protein contents in single cells by e.g., single-molecule fluorescence

counting [45] (Figure II-2-1). One of the many successful examples in the use of microfluidics in target

identification is the measurement of the number of epitope-tagged human β2-adrenergic receptors

(β2ARs), an important pharmacologic target in a number of airway and cardiovascular diseases [46].

Another interesting application of microfluidic devices in target identification is in ligand binding studies,

where microfluidics can minimize the interaction times, improving sensitivity and increasing throughput.

In this sense, a microfluidic platform was used to characterize DNA binding energy using four

eukaryotic transcription factors to predict their in vivo function [47].

b) For hit identification

b1) Compound generation: Developments in combinatorial chemistry have greatly enhanced our

ability to generate drug candidates [43]. Merging channel geometries, various reagents can be mixed

and biochemical reactions can be induced in a well-controlled and precise manner. Reduction of

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

31

reaction chamber’s size, and consequently, in the amount of reagents and time needed for reaction

empower this strategy. Mitchell et al. [48] developed a miniaturized-synthesis/total analysis system that

incorporated a chemical microprocessor with time-of-flight mass spectrometry (TOF-MS), which has

been used in the generation of compound libraries [48]. This system enabled real-time processing of

many reactions due to the integration of continuous-flow synthesis and provision for on-line analysis

within a microfabricated structure. In another example, Zhou et al. [49] developed a microfluidic pico-

array device for the synthesis and purification of oligonucleotides. This device consisted of fluidic

channels and 3,698 individual pico-reaction chambers for parallel synthesis of oligonucleotides [49].

Microfluidic systems can be also used to understand and assess natural drug candidates. Natural

products are undoubtedly one of the major sources of new chemical entities, with an increase in world

market growth rate [50]. Natural molecules are “evolutionarily optimized” as drug-like molecules and

remain one of the best sources of drug leads [50]. To identify the physiological action of a natural

compound out of a complex array of constituent molecules in the source material, microfluidic devices

can also be used. Gramowski et al. [51] applied multielectrode microchips for the screening of herbal

medicines. The screening of complex mixtures of neuroactive substances was assessed in the chip,

where changes in the spontaneous activity of cultured networks of primary cortical neurons were

quantified to evaluate the action of the drugs [51].

b2) Compounds screening: HTS is considered one of the most important methods for identifying

“hit” compounds [52]. Over the last decades, pharmaceutical companies established HTS as a major

tool to screen the properties of new chemical entities. A significant number of microfluidic technologies

can now be used to enable HTS studies, including multiplexed systems, microwell arrays, plug-based

methods and gradient-generating devices [53]. Khademhosseini et al. [54] studied the response of

multiple cell types (e.g. hepatocytes, fibroblasts, and embryonic stem cells) to different compounds by

seeding cells within microwells that were integrated within an array of reversibly sealed microfluidic

channels (Figure II-2). The ability to position many cell types on a single chip could be useful for

studying the effects of a series of compounds on different cell types. Concentration gradients also have

a great role in drug screening and cell-based studies. A gradient-generating microfluidic device can

generate a certain range of doses and thereby control different cell behaviors at the same time. By

using gradients, cells can be stimulated with controlled temporal and spatial resolution to study the

effects of drug concentration on chemotaxis. Occhetta et al. [55] introduced a branched channel to

generate a gradient and culture 3D micro-masses of adult hBM-MSCs under continuous and controlled

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

32

laminar flow perfusion [55]. Moreover, the platform allowed for conditioning the cellular micro-

aggregates with different combinations of growth factors, morphogens or drug molecules, in a high-

throughput fashion.

Figure II-2 - Microfluidics in drug discovery. 1) On-chip reaction. (A) Schematic of the on-chip reaction process. (B) Time

frames of a video that correspond to the steps described in A. A concentrated solution of crystal violet (a pH

indicator) in 0.1 N sodium hydroxide solution is loaded into the microchamber with the three-state valve half-

open; loading an equal amount of 1 N HCl into the chamber turns the solution from blue to yellow. Reprinted

with permission from [45]. Copyright © 2004 The National Academy of Sciences; 2) Schematic diagram of

reversible sealing of microfluidic arrays onto microwell-patterned substrates to fabricate multiphenotype cell

arrays. Initially, a PDMS microfluidic mold in the form of an array of microchannels was aligned on an array

of microwells. Each cell type was flown through an independent channel. The cells docked onto the

corresponding micro-wells, which resulted in a patterned array of cells. To deliver multiple solutions, the

PDMS microfluidic mold was removed and replaced with another mold, which was placed orthogonally to

create multiphenotype cell arrays inside each microchannel. Reprinted with permission from [54]. Copyright

© 2018 Copyright Clearance Center, Inc.

II-4. ORGAN-ON-A-CHIP AND DRUG DISCOVERY

The combination of microfluidics with human cells, together with developments in areas such as

stem cell research, regenerative medicine, biomaterials, and tissue engineering, have resulted into the

development of new generation of 3D in vitro models that better mimic in vitro human tissues and

organs. These new models are coined “organs-on-chip” devices and reproduce the functional unit of a

human organ and tissue. Therefore, organs-on-a-chip provide not only the biological relevance but also

the needed high-throughput characteristics needed in drug discovery applications. In the following, we

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

33

provide an overview of the developed technologies of organ-on-chip devices and their role on drug

testing and discovery.

II-4.1. Lung–on-a-chip

The seminal paper by Huh et al. [56] described the development of a biomimetic lung-on-a-chip

model capable to reconstitute organ-level lung functions to create a model pathologic pulmonary edema

(Figure II-3-1). In this model, the authors cultured two types of human lung cells (epithelial and

endothelial primary lung cells) in stacked parallel microchannels separated by a (porous) PDMS

membrane. Alike the human lung, the upper alveolar channel was filled with air, whereas the lower

microvascular channel was filled with liquid. Stretch was cyclically applied to the membrane by

additional vacuum actuator channels to mimic the physiological breathing movements of the lung. After

adding interleukin-2 (IL-2) to the microvascular channel, the fluid started to leak into the air

compartment, mimicking pulmonary edema. The authors tested this pulmonary disease model against

the pharmacological agent GSK2193874. This compound blocks ion channels activated by mechanical

strain, and inhibited leakage of fluid to the air compartment, suggesting it would be a viable option for

patients with pulmonary edema. The next step will be to hook this lung up to other chip-based organs

with the goal of being able to rapidly screen many drugs across many organs [56].

Jain et al. [57] developed another lung-on-a-chip with primary lung alveolar and endothelial cells.

This new chip design enabled whole human blood to be perfused through the vascular channel without

producing thrombus formation, while allowing high-resolution, real-time analysis of interactions between

human blood cells and endothelial cells in an in vivo-like context. Using this technology and, in

combination with novel analytical tools for quantitation of dynamic platelet-endothelial interactions and

clot formation, the authors demonstrated a key role of the epithelium in inflammation-driven vascular

thrombosis during lipopolysaccharide endotoxin (LPS)-induced acute lung injury. The study showed that

the model can be used to evaluate in vitro cytoprotective effects of Protease-activated Receptor-1 (PAR-

1), a potential therapeutic with anti-thrombotic and anti-inflammatory activities [57].

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

34

II-4.2. Liver-on-a-chip

The liver is one of the most relevant organs involved in maintaining physiological homeostasis. In

vitro models that could recapitulate functional liver tissue could have an enormous impact in the

pharmaceutical field, allowing for toxicity assessment of new drugs [39]. To develop such models,

hepatocytes with membrane polarity and functional bile canaliculi are essential [58]. Lee et al. [59]

created a biologically-inspired artificial liver sinusoid with a microfluidic endothelial-like barrier having

mass transport properties similar to the liver acinus (Figure II-3-2). This unit consisted of a cord of

hepatocytes (50 µm x 30 µm x 500 µm) fed by diffusion of nutrients across the microfluidic

endothelial-like barrier from a convective transport vessel. Combining photolithography-based

microfluidic technologies and primary cell culture, the team was able to assess liver toxicity of

diclofenac, since it has long been implicated in drug-related liver damage [59].

Following Lee’s work, Nakao et al. [58] presented a microfluidic device comprising an hepatic cord-

like structure model with primary rat hepatocytes, which was proven functional. Computer simulation

results showed that the flow velocity in the medium (1 mm∕s), matched the respective blood flow

velocity in vivo. Moreover, when 5-(and-6)-carboxy-2′,7′-dichloro-fluorescein diacetate (CDFDA) was

injected into the device, it was readily absorbed by the cells and metabolized into CDF by esterase. The

metabolites were actively excreted into bile canaliculi by the MRP2 protein [58].

II-4.3. Kidney-on-a-chip

Standard photolithography technology was used by Jang et al. [60] to develop a simple multi-layer

microfluidic device by integrating a PDMS microfluidic channel and a porous polyester membrane

substrate (0.4 μm pore size; 10 μm thickness) to culture primary rat inner medullary collecting duct

(IMCD) cells [60]. In order to evaluate molecular transport in renal tubule cells, water and sodium

uptake was measured after hormonal stimulations of the IMCD cells using vasopressin and aldosterone.

The cells were subjected to a fluidic shear stress of 1 dyn/cm2 for the time period of 5 h on the porous

membrane, which was found to be sufficient for the IMCD cells to enhance cell polarization and

rearrange cytoskeleton and cell junctions [60].

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

35

II-4.4. Heart-on-a-chip

Agarwal et al. [61] developed anisotropic cardiac microtissues recapitulating the laminar

architecture of the heart ventricle, which were engineered on cantilevers. The sub-millimeter-sized thin

film cantilevers of soft elastomers were fabricated by a laser-based process (Figure II-3-3) [61]. They

used the microdevice seeded with cardiac myocytes from mice to test the positive inotropic effect of

isoproterenol on cardiac contractility at dosages ranging from 1 nM to 100 μM. Tissues were exposed

to different drug concentrations with the advantage of complete washout between dosages, leading to a

fast collection of data from just one chip [61].

II-4.5. Bone marrow-on-a-chip

Torisawa et al. [62] developed a tissue-engineered bone marrow model that is able to retain

hematopoietic stem and progenitor cells in in vivo-like conditions for one week. In this example, the

bone marrow is first generated in mice, and then explanted as a whole and maintained in vitro within a

microfluidic device fabricated using standard photolithography (Figure II-3-4). The obtained results show

that this model mimics the complex tissue-level responses to Ƴ-radiation toxicity normally observed in

vivo, as well as to a therapeutic counter-measure agent (G-CSF) that is known to accelerate recovery

from radiation-induced toxicity in cancer patients [62]. Moreover, this research sheds light on the fact

that it is crucial to keep the whole hematopoietic niche to obtain an in vivo-like response, rather than

culturing particular cell types [62].

II-4.6. Intestine-on-a-chip

When discussing drug development, the small intestine is crucial, given that orally administered

drugs are mainly absorbed in the small intestine [63]. There they diffuse across a mucous layer

covering an epithelial cell layer lining the intestinal wall. Therefore, it is logical that drugs and chemicals

are tested on intestinal cells and tissues to assess the pharmacodynamics and pharmacokinetics of

drugs, such as absorption, distribution, metabolism, elimination and toxicity (ADMET).

Having this in mind, several groups started developing in vitro models for absorption and

metabolism studies. Realistic models should include certain key components such as cellular (i.e.

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

36

enterocytes, goblet and vascular endothelial cells), structural (i.e. villi and,mucus) and dynamic (i.e.

peristalsis) features. Kimura et al. [64] engineered an intestinal model with a membrane and vascular

flow simulating the epithelial barrier and the epithelial-endothelial interface. The microfluidic model

comprised optical fiber inserts for on-line and real-time pharmacokinetic measurements to detect the

amount of fluorescence of rhodamine-123 in culture, working as a tracer substrate to study diffusion

[64].

Figure II-3 - Organs-on-chip models: 1) A microengineered lung-on-a-chip model of human pulmonary edema. (A) IL-2

therapy is associated with vascular leakage. (B) IL-2–induced pulmonary edema is modeled in the lung-on-a-

chip. The top or “air” portion is the alveolar channel; the bottom or “liquid” portion is the vascular channel.

Scale bar represents 200 μm. (C) Endothelial exposure to IL-2 (1000 U/mL) causes liquid from the lower,

microvascular channel to leak into the upper, alveolar chamber. Scale bars represent 200 μm. (D) During IL-

2 treatment, prothrombin (100 μg/mL) and fluorescently labeled fibrinogen (2 mg/mL) introduced into the

microvascular channel form fluorescent fibrin clots (white) over the course of 4 days. Scale bar represents

200 μm. (E) A fluorescence confocal microscopic image shows that the fibrin deposits (red) in (D) are found

on the upper surface of the alveolar epithelium (green). Scale bar represents 50 μm. (F) The clots in (D) and

(E) are highly fibrous networks. Scale bar represents 5 μm. Reprinted with permission from [56]. Copyright

© 2012, American Association for the Advancement of Science. 2) Microfluidic liver-on-a-chip model. (a) The

liver receives blood flow from the hepatic artery and the portal vein, which carries xenobiotics such as

ingested drugs from the small intestine to be metabolized. (b) Schematic of the hepatocyte

microenvironment in liver tissue. (c) The sinusoid space is bordered by a sheet of highly fenestrated

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

37

endothelial cells. (d) The single microfluidic sinusoid had three fluid terminals: a flow inlet, a flow outlet, and

a cell inlet. Scale bar represents 20 μm. Reprinted with permission from [59]. Copyright © 2007 Wiley

Periodicals, Inc. 3) Higher-throughput heart-on–a-chip model. (a) Schematic of the fabrication process. (b)

Image of the engraving laser processing of Muscular Thin Films (MTFs) in the 18 mm diameter chip. (c)

Exploded view of the conception and assembly of the fluidic device which fits the chip. (d) Image of an actual

device in action. Reprinted with permission from [61]. Copyright © 2018 Royal Society of Chemistry. 4) In

vivo bone marrow-on-a-chip model. (a) Workflow to generate a bone marrow-on-a-chip system (b) PDMS

device containing bone-inducing materials in its central cylindrical chamber before implantation (top). (c)

Low- (left) and high-magnification views (right) of histological hematoxylin-and-eosin–stained sections of the

bone marrow formed in the PDMS device with two openings (top) or one lower opening (center) at 8 weeks.

Reprinted with permission from [62]. Copyright © 2014, Springer Nature.

II-4.7. Tumor-on-a-chip

Cancer remains one of the main causes of death, despite the enormous efforts to cure the disease

[65, 66]. Solid evidences have shown this key role of the tumor microenvironment (TME) in the onset

and progression of the disease. A better understanding of the tumor microenvironment is therefore

needed to improve our knowledge about the mechanistic determinants of the disease. However, the

TME is extremely complex from a structural, cellular, and dynamic perspective, and therefore, very

difficult to investigate in vitro and in vivo. The emerging tumor-on-a-chip technology can be of extreme

usefulness in these processes. as well as screening platforms to evaluate the efficiency of anti-cancer

drugs [67, 68].

In an approach to better understand breast cancer metastasis to bone, Jeon et al. [69] developed

a microfluidic model of a human vascularized organ-specific microenvironment, which was used to

explore and tune the extravasation process of metastatic tumor cells [69]. The experiment comprised

an organ-specific model that enabled the study of human metastatic breast cancer cell extravasation

within a perfusable human microvascularized bone-mimicking microenvironment. The microfluidic

model was characterized by anastomoses with the lateral media channels. MDA-MB-231 breast cancer

cells were introduced and extravasation events monitored. The platform was used to understand the

anti-metastatic role of adenosine in the human breast cancer metastasis to bone [69].

In its turn, Bischel et al. [70] integrated a microfluidic co-culture platform with a multi-photon

imaging-based technique to determine phenotypic cell behavior and flavin adenine dinucleotide (FAD)

fluorescence intensity and fluorescence lifetime simultaneously in the same cell. This platform

combined two independent assays normally performed with two different cell populations into a single

device, allowing researchers to simultaneously assess both phenotypic cell behavior and enzyme activity

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

38

[70]. The metastasis towards bone was mimicked by differentiating MC3T3-E1 cells (osteoblast

precursor cell line) towards osteoblast lineage and seeded on the sidewall of a microchannel. Next, a

collagen I hydrogel coating was applied as extracellular matrix, either Lymph Node Carcinome of the

Prostate cells (LNCaP) or (Bone metastatic LNCaP-derivative C4-2B) C4-2B cells were seeded into the

channel. By engineering this cancer-bone microenvironment, it was demonstrated that the cross-talk

between a bone metastatic prostate cancer cell line, C4-2B, and bone stromal cells, MC3T3-E1,

increased the invasive behavior of the C4-2Bs, possibly by an increase in reactive oxygen species (ROS)-

producing N1-acetyl polyamine oxidase (APAO) activity [70].

Sung et al. [71] presented a microscale cell culture analog (mCCA), which is basically a

microfluidic device based on pharmacokinetics-pharmacodynamics (PK-PD) mathematical models. In

this model, multiple cell culture chambers were connected with fluidic channels to mimic multi-organ

interactions and test drug toxicity in a pharmacokinetic-based manner [71]. Three cell lines representing

the liver, a tumor and bone marrow were cultured in the three-chamber microCCA to test the toxicity of

an anticancer drug, 5-fluorouracil (5-FU). The result was analyzed with the PK-PD model of the device,

and compared with the result of the static conditions (cell culture flasks). Each cell type exhibited

differential responses to 5-FU, and the responses in the microfluidic environment were different from

those in the static environment. The authors showed a relatively new approach based on the

combination of mathematical modeling and an in vitro experimentation, with improved predictability for

testing drug toxicity.

More recently, Wong et al. [72] exploited a 2.4 × 2.4 cm PDMS-based microfluidic chip to generate

droplets and perform drug screening against suspended and adherent cancer cell lines, as well as cells

dissociated from primary tumors of human patients. Bortezomib and Vorinostat were chosen as target

drugs for leukemia in conjunction with Jurkat cells, whereas Cisplatin and Epirubicin were chosen as

target drugs for breast cancer in conjunction with MDA-MB-231 cells. Cell viability was measured by

ethidium homodimer-1 staining, enabling the capture of single cell drug response, without

compromising population analysis. This method provided a versatile and rapid drug screening method

using cells in different states [72].

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

39

II-5. CONCLUSION AND FUTURE TRENDS

New micro-engineering-based strategies have significantly improved the drug development process.

Microfluidics makes possible to assess their therapeutic activity by enabling high-throughput functional

readouts; to conduct high-throughput screens for drug discovery in realistic engineered 3D culture

microenvironments; to develop organ-on-a-chip platforms to screen those drugs, which would probably

fail during the expensive animal testing and human clinical trials later in the drug discovery process

pipeline. Critical challenges remain in making these technological advances practically feasible for real-

world pharmaceutical discovery. Thus, the optimal solution for drug discovery may rely on sufficiently

complex and realistic technology, while being simple enough to allow for assay readouts, reliability and

operational simplicity. In addition, mathematical and computational modeling will guide the design of

the next generation of organ-on-chip devices, and bio-sensing will enable the real-time and in situ

monitoring of the cellular metabolic responses.

Regarding biomaterials, in vitro platforms are just a scientific attempt at mimicking the natural

system. It is therefore important to make the right choice of the function to be mimicked, as this choice

will eventually determine whether a system is considered valuable or not. Because of this, there is now

a huge demand for new and tunable biomaterials to interface with microfluidic chips for cell biological

studies.

However, the reality is that despite rapid development of expertise in designing, fabricating and

operating microfluidic systems for nearly twenty years now, no commercially available drugs have been

discovered as a unique result of microfluidic technologies, to the best of our knowledge.

Griffith, a world-class scientist in organ-on-a-chip research said that more questions would be

answered if Pharma collaborated more with engineers, as Pharma knows what problems it needs to

solve, but engineers know how to build chips. Therefore, success in this endeavor must start with

alliances between the fields of pharmaceutical industry, engineering, medicine, biology and

computational biology. Maybe that is our future answer.

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

40

II-6. REFERENCES

1. Mullard, A., 2010 FDA drug approvals. Nat Rev Drug Discov, 2011. 10(2): p. 82-5. 2. Hughes, J.P., et al., Principles of early drug discovery. British Journal of Pharmacology, 2011. 162(6): p. 1239-1249. 3. Hay, M., et al., Clinical development success rates for investigational drugs. Nature Biotechnology, 2014. 32: p. 40. 4. Paul, S.M., et al., How to improve R&D productivity: the pharmaceutical industry's grand challenge. Nat Rev Drug Discov, 2010. 9(3): p. 203-14. 5. Mak, I.W., N. Evaniew, and M. Ghert, Lost in translation: animal models and clinical trials in cancer treatment. Am J Transl Res, 2014. 6(2): p. 114-8. 6. Szymański, P., M. Markowicz, and E. Mikiciuk-Olasik, Adaptation of High-Throughput Screening in Drug Discovery—Toxicological Screening Tests. International Journal of Molecular Sciences, 2012. 13(1): p. 427-452. 7. Zheng, X.T., et al., On-chip investigation of cell-drug interactions. Adv Drug Deliv Rev, 2013. 65(11-12): p. 1556-74. 8. Khanna, I., Drug discovery in pharmaceutical industry: productivity challenges and trends. Drug Discovery Today, 2012. 17(19): p. 1088-1102. 9. Chandrasekaran, A., M. Abduljawad, and C. Moraes, Have microfluidics delivered for drug discovery? Expert Opinion on Drug Discovery, 2016. 11(8): p. 745-748. 10. Whitesides, G.M., The origins and the future of microfluidics. Nature, 2006. 442: p. 368. 11. Bettinger, C.J. and J.T. Borenstein, Biomaterials-based microfluidics for engineered tissue constructs. Soft Matter, 2010. 6(20): p. 4999-5015. 12. Dittrich, P.S. and A. Manz, Lab-on-a-chip: microfluidics in drug discovery. Nature Reviews Drug Discovery, 2006. 5: p. 210. 13. Tsao, C.-W., Polymer Microfluidics: Simple, Low-Cost Fabrication Process Bridging Academic Lab Research to Commercialized Production. Micromachines, 2016. 7(12). 14. FluidicMEMS. List of microfluidics companies. 2018; Available from: http://fluidicmems.com/list-of-microfluidics-lab-on-a-chip-and-biomems-companies/. 15. Tian, C., et al., Recent advances in microfluidic technologies for organ-on-a-chip. TrAC Trends in Analytical Chemistry, 2019. 16. Shantanu, P., et al., Polymeric Biomaterials for In Vitro Cancer Tissue Engineering and Drug Testing Applications. Tissue Engineering Part B: Reviews, 2016. 22(6): p. 470-484. 17. Carvalho, M.R., et al., Significance of Tissue-engineered 3D in-vitro models for cancer research and drug screening. Trends Biotechnol, 2015. 18. Kobel, S. and M.P. Lutolf, Biomaterials meet microfluidics: building the next generation of artificial niches. Current Opinion in Biotechnology, 2011. 22(5): p. 690-697. 19. Radhakrishnan, J., et al., Injectable and 3D Bioprinted Polysaccharide Hydrogels: From Cartilage to Osteochondral Tissue Engineering. Biomacromolecules, 2017. 18(1): p. 1-26. 20. Lai, J.H., et al., Stem cells catalyze cartilage formation by neonatal articular chondrocytes in 3D biomimetic hydrogels. 2013. 3: p. 3553. 21. Cui, X., et al., Direct human cartilage repair using three-dimensional bioprinting technology. Tissue Eng Part A, 2012. 18(11-12): p. 1304-12. 22. Bichara, D.A., et al., Osteochondral defect repair using a polyvinyl alcohol-polyacrylic acid (PVA-PAAc) hydrogel. Biomed Mater, 2014. 9(4): p. 045012.

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

41

23. de Girolamo, L., et al., Repair of osteochondral defects in the minipig model by OPF hydrogel loaded with adipose-derived mesenchymal stem cells. Regen Med, 2015. 10(2): p. 135-51. 24. Inagaki, Y., et al., Effects of culture on PAMPS/PDMAAm double-network gel on chondrogenic differentiation of mouse C3H10T1/2 cells: in vitro experimental study. BMC Musculoskelet Disord, 2014. 15: p. 320. 25. Zeng, L., et al., Effect of microcavitary alginate hydrogel with different pore sizes on chondrocyte culture for cartilage tissue engineering. Materials Science and Engineering: C, 2014. 34: p. 168-175. 26. Lutolf, M.P. and J.A. Hubbell, Synthetic biomaterials as instructive extracellular microenvironments for morphogenesis in tissue engineering. Nat Biotechnol, 2005. 23(1): p. 47-55. 27. Barata, D., C. van Blitterswijk, and P. Habibovic, High-throughput screening approaches and combinatorial development of biomaterials using microfluidics. Acta Biomaterialia, 2016. 34: p. 1-20. 28. Wan, J., Microfluidic-Based Synthesis of Hydrogel Particles for Cell Microencapsulation and Cell-Based Drug Delivery. Polymers, 2012. 4(2). 29. Baker, B.M., et al., Microfluidics embedded within extracellular matrix to define vascular architectures and pattern diffusive gradients. Lab Chip, 2013. 13(16): p. 3246-52. 30. Toh, Y.C., et al., A microfluidic 3D hepatocyte chip for drug toxicity testing. Lab Chip, 2009. 9(14): p. 2026-35. 31. Dinh, N.D., et al., Microfluidic construction of minimalistic neuronal co-cultures. Lab Chip, 2013. 13(7): p. 1402-12. 32. Si, G., et al., A parallel diffusion-based microfluidic device for bacterial chemotaxis analysis. Lab on a Chip, 2012. 12(7): p. 1389-1394. 33. Bita, C., et al., Recreating the perivascular niche ex vivo using a microfluidic approach. Biotechnology and Bioengineering, 2010. 107(6): p. 1020-1028. 34. Anguiano, M., et al., Characterization of three-dimensional cancer cell migration in mixed collagen-Matrigel scaffolds using microfluidics and image analysis. PLoS One, 2017. 12(2): p. e0171417. 35. Chaw, K.C., et al., Matrigel coated polydimethylsiloxane based microfluidic devices for studying metastatic and non-metastatic cancer cell invasion and migration. Biomed Microdevices, 2007. 9(4): p. 597-602. 36. Gao, D., et al., A microfluidic approach for anticancer drug analysis based on hydrogel encapsulated tumor cells. Anal Chim Acta, 2010. 665(1): p. 7-14. 37. Zhang, Y., Y. Yu, and I.T. Ozbolat, Direct Bioprinting of Vessel-Like Tubular Microfluidic Channels. Journal of Nanotechnology in Engineering and Medicine, 2013. 4(2): p. 0210011-0210017. 38. Hussey, G.S., J.L. Dziki, and S.F. Badylak, Extracellular matrix-based materials for regenerative medicine. Nature Reviews Materials, 2018. 39. Uygun, B.E., et al., Organ reengineering through development of a transplantable recellularized liver graft using decellularized liver matrix. Nat Med, 2010. 16(7): p. 814-20. 40. Choi, K.-H., et al., The chondrogenic differentiation of mesenchymal stem cells on an extracellular matrix scaffold derived from porcine chondrocytes. Biomaterials, 2010. 31(20): p. 5355-5365. 41. Lu, H., et al., Autologous extracellular matrix scaffolds for tissue engineering. Biomaterials, 2011. 32(10): p. 2489-99. 42. Jusoh, N., et al., Microfluidic vascularized bone tissue model with hydroxyapatite-incorporated extracellular matrix. Lab Chip, 2015. 15(20): p. 3984-8. 43. Kang, L., et al., Microfluidics for Drug Discovery and Development: From Target Selection to Product Lifecycle Management. Drug discovery today, 2008. 13(1-2): p. 1-13.

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

42

44. Barata, D., et al., Development of a shear stress-free microfluidic gradient generator capable of quantitatively analyzing single-cell morphology. Biomed Microdevices, 2017. 19(4): p. 81. 45. Wu, H., A. Wheeler, and R.N. Zare, Chemical cytometry on a picoliter-scale integrated microfluidic chip. Proc Natl Acad Sci U S A, 2004. 101(35): p. 12809-13. 46. Huang, B., et al., Counting low-copy number proteins in a single cell. Science, 2007. 315(5808): p. 81-4. 47. Maerkl, S.J. and S.R. Quake, A systems approach to measuring the binding energy landscapes of transcription factors. Science, 2007. 315(5809): p. 233-7. 48. Mitchell, M.C., V. Spikmans, and A.J.d. Mello, Microchip-based synthesis and analysis: Control of multicomponent reaction products and intermediates. Analyst, 2001. 126(1): p. 24-27. 49. Zhou, X., et al., Microfluidic PicoArray synthesis of oligodeoxynucleotides and simultaneous assembling of multiple DNA sequences. Nucleic Acids Res, 2004. 32(18): p. 5409-17. 50. Shen, B., A New Golden Age of Natural Products Drug Discovery. Cell, 2015. 163(6): p. 1297-1300. 51. Gramowski, A., et al., Functional screening of traditional antidepressants with primary cortical neuronal networks grown on multielectrode neurochips. Eur J Neurosci, 2006. 24(2): p. 455-65. 52. Chandrasekaran, A., M. Abduljawad, and C. Moraes, Have microfluidics delivered for drug discovery? Expert Opin Drug Discov, 2016. 11(8): p. 745-8. 53. Du, G., Q. Fang, and J.M.J. den Toonder, Microfluidics for cell-based high throughput screening platforms—A review. Analytica Chimica Acta, 2016. 903: p. 36-50. 54. Khademhosseini, A., et al., Cell docking inside microwells within reversibly sealed microfluidic channels for fabricating multiphenotype cell arrays. Lab Chip, 2005. 5(12): p. 1380-6. 55. Occhetta, P., et al., High-Throughput Microfluidic Platform for 3D Cultures of Mesenchymal Stem Cells, Towards Engineering Developmental Processes. Scientific Reports, 2015. 5: p. 10288. 56. Huh, D., et al., A human disease model of drug toxicity-induced pulmonary edema in a lung-on-a-chip microdevice. Sci Transl Med, 2012. 4(159): p. 159ra147. 57. A, J., et al., Primary Human Lung Alveolus‐on‐a‐chip Model of Intravascular Thrombosis for Assessment of Therapeutics. Clinical Pharmacology & Therapeutics, 2018. 103(2): p. 332-340. 58. Nakao, Y., et al., Bile canaliculi formation by aligning rat primary hepatocytes in a microfluidic device. Biomicrofluidics, 2011. 5(2): p. 22212. 59. Lee, P.J., P.J. Hung, and L.P. Lee, An artificial liver sinusoid with a microfluidic endothelial-like barrier for primary hepatocyte culture. Biotechnol Bioeng, 2007. 97(5): p. 1340-6. 60. Jang, K.J. and K.Y. Suh, A multi-layer microfluidic device for efficient culture and analysis of renal tubular cells. Lab Chip, 2010. 10(1): p. 36-42. 61. Agarwal, A., et al., Microfluidic heart on a chip for higher throughput pharmacological studies. Lab on a chip, 2013. 13(18): p. 3599-3608. 62. Torisawa, Y.-s., et al., Bone marrow–on–a–chip replicates hematopoietic niche physiology in vitro. Nature Methods, 2014. 11: p. 663. 63. Fasinu, P., et al., Diverse approaches for the enhancement of oral drug bioavailability. Biopharm Drug Dispos, 2011. 32(4): p. 185-209. 64. Kimura, H., et al., An integrated microfluidic system for long-term perfusion culture and on-line monitoring of intestinal tissue models. Lab Chip, 2008. 8(5): p. 741-6. 65. Stroock, A.D. and C. Fischbach, Microfluidic culture models of tumor angiogenesis. Tissue Eng Part A, 2010. 16(7): p. 2143-6. 66. Buchanan, C. and M.N. Rylander, Microfluidic culture models to study the hydrodynamics of tumor progression and therapeutic response. Biotechnol Bioeng, 2013. 110(8): p. 2063-72.

Chapter II – Biomaterials and Microfluidics for Drug Discovery and Development

43

67. Carvalho, M.R., et al., Anti-Cancer Drug Validation: the Contribution of Tissue Engineered Models. Stem Cell Rev, 2017. 68. Carvalho, M.R., et al., Evaluating Biomaterial- and Microfluidic-Based 3D Tumor Models. Trends in Biotechnology. 33(11): p. 667-678. 69. Jeon, J.S., et al., Human 3D vascularized organotypic microfluidic assays to study breast cancer cell extravasation. Proceedings of the National Academy of Sciences, 2015. 112(1): p. 214-219. 70. Bischel, L.L., et al., A microfluidic coculture and multiphoton FAD analysis assay provides insight into the influence of the bone microenvironment on prostate cancer cells. Integrative Biology, 2014. 6(6): p. 627-635. 71. Sung, J.H., C. Kam, and M.L. Shuler, A microfluidic device for a pharmacokinetic-pharmacodynamic (PK-PD) model on a chip. Lab Chip, 2010. 10(4): p. 446-55. 72. Wong, A.H.-H., et al., Drug screening of cancer cell lines and human primary tumors using droplet microfluidics. Scientific Reports, 2017. 7(1): p. 9109.

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

44

Chapter III

Dendrimer Nanoparticles in

CRC applications

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

45

Chapter III

Chapter III - Dendrimer Nanoparticles in CRC applications

ABSTRACT

Cancer nanotechnology is a prolific field of research, where nanotools are employed to diagnose and

treat cancer with unprecedented precision. Targeted drug delivery is fundamental for more efficient

cancer treatments. For this, nanoparticles have been extensively used during the last years in order to

improve the specificity, selectivity and controlled release of drug delivery. It holds potential in minimizing

systemic toxicity through the development of functionalized particles for targeted treatment. Among all

the type of nanoparticles, dendrimers display several advantages, which make them ideal candidates

for improved and targeted drug delivery in cancer research. Dendrimers can transport large amount of

drug into specific areas. In addition, they can be employed for monitoring the progress of the treatment

process, with an unprecedented theranostic capability. They can help to limit the side effects of

anticancer drugs and limit their action toward specific regions. This chapter provides a brief overview on

the types of dendrimer nanoparticles, their synthesis routes, as well as the preferred employed

strategies for producing drug-loaded/functionalized NP’s for cancer therapy in the last years. Special

emphasis is given in colorectal cancer, which has shown a dramatic increase in its incidence during the

last years, especially in industrialized countries.

Keywords: Colorectal cancer; Dendrimer nanoparticles; Targeting.

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

46

III-1. INTRODUCTION

Nanotechnology is an emerging field of research that is expected to play a definitive role in

biomedicine in the near future, particularly for the targeted delivery of drugs at pathological sites inside

the human body [1]. Nanoparticles (NPs) can be defined as particulate matter ranging between 1 and

100 nm in diameter, with a high surface area-to-volume ratio, and specialized surface characteristics [2-

4]. The use of NPs as drug carriers in oncology started in 1986, when it was reported that NPs showed

a tendency to accumulate in tumoral tissues [5]. This passive accumulation, also known as passive or

primary targeting, is known as “Enhanced Permeability and Retention” (EPR) effect, and is one of the

milestones of cancer treatment using NPs. Therefore, the use of NPs for cancer therapeutics holds very

promising due to their high specificity and accumulation in tumor sites, as well as their long blood

circulation time (Figure III-1).

However, one of the major problems related to cancer treatment is its intrinsic anti-cancer drug

resistance, which can appear prior to chemotherapy as well as acquired resistance due to drug

treatment [6]. The selectivity improvement yielded by NPs resulted in a great enhancement in the

efficacy of the transported drug, while the occurrence of side effects in the patient was reduced. Hence,

targeted nanomedicine offers innovative therapeutic strategies to overcome the various limitations of

conventional chemotherapy, such as drug resistance, enabling enhanced selectivity, and early and

more precise cancer diagnosis [7]. Moreover, it is possible to incorporate targeting moieties selective for

cancer cell biomarkers, which improves even more the selectivity and specificity of the treatment [8].

The main types of targeting moieties that are used for decorating and targeting NPs are: i) Antibodies

[9]; ii) Peptide-based targeting [10]; iii) Small molecule-based targeting [11]; and iv) Aptamer-based

targeting [12].

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

47

Figure III-1 - Rationale for nanomedicine approach in cancer-therapy. Reprinted with permission from [1].

The discovery of non-biodegradable nanoparticles including micelles, nanogels, liposomes,

nanoemulsions, polymeric NPs, gold NPs and magnetic NPs, as agents in nano-drug delivery and

imaging at pathological sites has enhanced delivery at lower doses and increased aqueous solubility

and bioavailability of the drug with reduced side effects [13]. Nevertheless, the clinical application of

these NPs to CRC therapy remains limited, as can be seen in Table III-1.

Table III-1 - Current nanotechnology applications in colorectal cancer clinical trials and status.

Study Title Identifier Intervention Status

Targeted polymeric nanoparticles loaded with cetuximab and decorated with somatostatin

analogue to colon cancer NCT03774680

Cetuximab NPs/ Oral approved

anticancer drug. Recruiting

Targeted silica nanoparticles for real time image-guided intraoperative mapping of nodal

metastasis NCT02106598

Fluorescent cRGDY-PEG-Cy5.5-C dots

Recruiting

TKM 080,301 (Lipid nanoparticles containing siRNA against the PLK1 gene product) in patients with colorectal, pancreas, gastric,

breast, ovarian and esophageal cancer with hepatic/ TKM 080,301 for primary or

secondary liver cancer

NCT01437007 Drug: TKM-080,301 Completed

Pharmacokinetic, safety and efficacy study of nanoparticle paclitaxel in patients with peritoneal cancer/ A Phase I study of

intraperitoneal nanoparticle paclitaxel in patients with peritoneal malignancies

NCT00666991 Nanoparticulate

paclitaxel Completed

Neoadjuvant chemoradiotherapy with CRLX-101 and capecitabine for rectal cancer

NCT02010567 CRLX101/

Capecitabine/ Radiotherapy

Active/ Not recruiting

Liposomal irinotecan, fluorouracil, leucovorin calcium and rucaparib in treating patients

with metastatic pancreatic, colorectal,

NCT03337087 5-FU/ Leucovorin

calcium/ liposomal irinotecan, rucaparib

Recruiting

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

48

gastroesophageal or biliary cancer

Moreover, as can be seen from Table III-1, none of the ongoing clinical trials using nanotechnology in

colorectal cancer comprises the use of dendrimer nanoparticles. After literature research, it was found

that there is only one clinical trial using dendrimers, with application in inoperable liver cancers. This

dendrimer consists of poly-L-lysine dendrimer as nanovector mixed with complex of 188-Rhenium-ligand

(nitro-imidazole-methyl-1,2,3-triazol-methyl-di-[2-pycolyl] amine) [14]. Although a very popular type of

nanoparticles, inclusively used in clinical trials for other pathologies such as breast cancer and HIV

prevention (VivaGel®), scientific advances are needed for dendrimer applications in the clinical setting.

III-2. DENDRIMER NANOPARTICLES

III-2.1. Properties

Among the innumerous types of NPs available, dendrimers offer multiple advantages. They are

highly branched polymers, and therefore specific moieties and drugs can be easily conjugated and

encapsulated. Herein, we focused on polyamidoamine (PAMAM) dendrimer NPs. PAMAM dendrimers

are the most common class of dendrimers, suitable for many tissue engineering and regenerative

medicine (TERM), materials science, and biotechnology applications [15]. They contain an inner alkyl-

diamine core and a peripheral shell made of tertiary amine branches [16]. They have a well-designed

branching architecture with abundant terminal groups [17]. Therefore, the high level of control over

dendritic architectures makes dendrimers ideal carriers in biomedical applications [18]. In addition, the

toxicity of dendrimers mainly comes from the high cationic charge density in the periphery, where

charges interact with biological cell membrane, resulting in membrane disruption, hurdle that can be

easily overcome by surface modification. Dendrimers have the advantages of being biocompatible upon

modification and easily eliminated from the body [19] through the kidneys along the same metabolic

pathways taken by folate, peptides and antibodies [20-22].

Since dendrimers possess high density of surface functional molecules, they are easily conjugated

with several targeting agents for selective delivery of chemotherapeutics to the tumor tissue.

Furthermore, they contain internal cavities for macromolecule encapsulation, being able to transport

highly hydrophobic drugs (Figure III-2) [23].

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

49

Figure III-2 - Dendrimer branched architecture, representing the increasing generations (G). Reprinted with permission from

[24].

III-2.2. Synthesis and types of Dendrimers

Dendrimers have been traditionally synthesized by two major routes: the divergent method,

introduced by Tomalia [17], and the convergent one, developed by Hawker and Frechet [25]. In the first

method (divergent), the final molecule grows radially from a core by the sequential addition of layers of

monomers, each layer constituting a new generation (G). The number of surface groups multiplies

according to the functionalities in each monomer ramification [26]. It is important that every step of the

reaction is fully completed before the addition of a new generation to avoid defects in the branches. One

of the main advantages of this approach is that in the final step of the reaction, the surface of the

dendrimer can be easily modified with desired functional groups. Moreover, it is a fast synthesis which

allows the preparation of large dendrimers. The main drawbacks of this approach is the extensive

purification it is required, since the final product and the intermediate reactants have similar molecular

weights, charge, and polarity [27]. Also, the higher the generation, the greater the chances are of

having branching defects, since the presence of bulky branches creates difficulties in the coupling of

new ones. Despite these obstacles, the advantages of this strategy have made it the main route for

dendrimer production.

In a reverse way from the divergent synthesis, dendrimers can also be synthesized starting from

the surface towards de inner core. The growth of the molecule starts from the ends of the chain,

beginning by integrating the various branching points with other monomers that will constitute the

dendrimers. Finally, these branches are attached to a central core when they reach the desired

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

50

generation size [28]. In contrast to the divergent growth, this method permits easier purification due to

bigger differences between the final products and the initial reagents. Other advantages include higher

monodispersity for low generations and fewer branch defects. The main downsides are low yield and

difficulties in obtaining higher generations due to steric interferences when the branches are connected

to the core [29].

Regarding the available types of dendrimers, Poly(propylene imine) (PPI) dendrimers were the first

ones to be reported by in the 70’s. Together with PAMAM, they represent the most widely studied type.

PPI dendrimers can be based on a 1,4-diaminobutane (DAB) core, but also be synthesized from an

ethylenediamine nucleus and other core molecules by a double Michael addition reaction. PPI

monomers are used as branching units. Hence their interior contains various tertiary tris-propylene

amines, and they form full generations with primary amines as surface ends [30].

Another common type of dendrimers are the Poly-l-lysine (PLL) dendrimers, mostly used as gene

carriers due to their excellent condensation with oligonucleotides. Among their favorable characteristics

are good biocompatibility, water solubility, biodegradability, and flexibility, similar to other dendrimers.

With peptide bonds in their structures, both their core and branching units are commonly based on the

amino acid lysine PLL dendrimers differ from the general concept of PAMAM and PPI dendrimers since

they are mostly asymmetrical [31]. However, they are still precise molecules, with a controlled number

of lysines branching out from the core, and terminal amine residues. The lysine in the terminal group of

PLL contains two primary amines that are frequently modified for better biological performance [32].

There are many more types of dendrimer nanoparticles, although less used. They can be classified

according to their shape, structure, branching, solubility, chirality and attachment, which can be seen in

Figure III-3.

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

51

Figure III-3 - Different types of dendrimers. Reprinted with permission from [33].

When compared to other dendritic NPs, PAMAM surpasses them due to the ease of its

preparation, desirable chemical and physical properties, surface functional groups, and comparatively

lower toxicity to other dendrimers [22].

A variety of dendrimers have been developed and used since the 1980s, but the ones derived from

polyamidoamine (PAMAM) are undeniably the most employed. They are generally, biocompatible, and

non-immunogenic systems, which favors their use in drug delivery. The core of PAMAM is most

commonly ethylenediamine, although more hydrophobic molecules such as diaminododecane,

diaminoexane, and diaminobutane may also be used [34].

Functionalization is an integral part of dendrimer multiple uses. Functionalization is the process of

incorporating multiple active sites in dendrimers in order to generate macromolecules with

multifunctional architecture. Functionalized dendrimers have at least six well defined nanoscale features

known as “Critical Nanoscale Design Parameters” (CNDPs). These parameters consist of size, shape,

surface chemistry, flexibility, rigidity, architecture and elemental composition [35]. These CNPDs may

be systematically employed to produce a wide range of new emerging properties that may be desirable

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

52

and critical for different clinical applications. The functionalization serves a variety of purposes, namely

to reduce cytotoxicity and enhance transepithelial transport [36], for interaction with coupling molecules

such as natural-based polymers [36], fluorescent probes [36], and an inner hydrophobic core where

other molecules can be trapped [37, 38]. Among the myriad possibilities, dendrimers can find

applications as delivery carriers of drugs and DNA (transfection) [39], imaging agents [37, 40, 41], and

scaffolds for TERM [42].

III-3. DENDRIMER APPLICATIONS IN CANCER RESEARCH

A strong body of evidence now suggests that NPs in the form of dendrimers may be of added value

in the future of oncology-related theranostics [43, 44]. Other than the flexibility for functionalization

using diverse ligands and its low-nanometer size, the introduction of stimuli responsive functionality on

dendrimers allows the release of payloads in response to specific triggers. These triggers could be

endogenous in nature (acid, enzyme, and redox potentials) or it could be applied externally (light and

temperature) [45-47]. One of these examples is the recent work developed by Nigam et al. [48] on

cervical cancer, where iron oxide NPs were modified with Generation 2 (G2) PAMAM dendrimers and

loaded with doxorubicin (DOX), therefore combining magnetic chemotherapy and hyperthermia on HeLa

cancer cells [48]. When exposed to alternating current magnetic field, results show enhanced cell death

as result of fatal synergistic contribution of DOX and high temperatures. Interestingly, the combinatorial

treatment reduced cancer cell viability from 100 % to 3.6 % [48].

In a unique strategy to fight lung cancer, Amreddy et al. [49] designed a targeted drug delivery

system to co-encapsulate functionally different agents, such as a combination of siRNA and

chemotherapy. For targeting purposes, folic acid was conjugated to PAMAM dendrimer NPs [49]. The

team developed a new NP system for co-delivery of siRNA against HuR mRNA (HuR siRNA) and cis-

diamine platinum (CDDP) to folate receptor-α (FRA) overexpressing H1299 lung cancer cells. Results

showed negligible toxicity towards non-cancerous MRC9 lung fibroblast cells, as well as the ability of

dendrimer nanoparticles for targeted co-delivery of siRNA and chemotherapy agents together in lung

cancer cells [49].

For breast cancer, one good example of dendrimer application is the work of Chittasupho et al.

[50]. Having in mind that breast tumors preferentially metastasize to the lung, bone and distant lymph

nodes (that secret high levels of CXCL12), the team hypothesized that targeted inhibition of CXCR4 in

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

53

breast cancer cells should suppress CXCR4-positive tumor cells toward secondary metastatic sites. The

efficacy of CXCR4 targeted dendrimers carrying doxociclin (LFC131-DOX-D4) on cellular binding,

cytotoxicity, and migration of BT-549-Luc and T47D breast cancer cells was investigated. Results

showed enhanced in vitro cellular toxicity as compared with non-targeted dendrimers. The modified

dendrimers exhibited remarkable reduced migration of BT-549-Luc breast cancer cells towards the used

chemoattractant. This report demonstrated the potential utility of LFC131-dendrimer conjugates for

breast cancer therapy and metastasis.

In the case of liver cancer, due to lack of cell surface biomarkers and highly metastatic nature,

early detection and targeted therapy of hepatocellular cellular carcinoma (HCC) is an unmet clinical

need [51]. Galactosamine (Gal) is among the few selective ligands used for targeting HCCs due to its

high binding affinity to asialoglycoprotein receptors (ASGPRs) overexpressed in HCC. In a recent work,

Yousef et al. [51] engineered a nanoscale G4 PAMAM dendrimer NP anchored to Gal and loaded with

the potent anticancer curcumin derivative (CDF) as a platform for targeted drug delivery to HCC.

Surprisingly, in an in vivo xenograft model, cytotoxicity assays in HCC cell lines showed that CDF was

more potent as a chemotherapeutic anticancer than the currently in use Doxorubicin, Sorafenib and

Cisplatin chemotherapeutic agents [51].

Glioblastoma is the most common type of malignant brain tumor and one of the deadliest cancers

[52]. It has been described that dendrimers have affinity to cross blood–brain barrier after systemic

administration [53]. Liu et al. [54] utilized a combined chemo- and gene-therapy approach for effective

glioma treatment and developed DOX-loaded dendrigraft poly-l-lysine (DGL) dendrimer surface modified

with TNF (tumor necrosis factor) related apoptosis-inducing ligand (TRAIL) for the tumor targeting

specifically of T7 peptide, a TfR-specific peptide. This approach was based on the knowledge that DOX

increased the anticancer effect of TRAIL by regulating the expression of death receptors, as well as

stimulation of apoptotic pathways [54, 55].

Other than targeting and locally releasing drugs, advances in nanotechnology have enabled the

development of different dendrimers that can be used as diagnostic aids in several imaging

applications. In particular, diagnostic applications including magnetic resonance imaging (MRI) [56]. For

example, Langereis et al. [57] reported the synthesis of 5–6-nm gadolinium-diethylene triamine penta-

acetic acid (Gd-DTPA)-terminated poly(propylene imine) (PPI) dendrimers as MRI agent. Talanov et al.

[58] reported a PAMAM dendrimer-based nanoprobe with dual MR and fluorescence (Fl) modalities. Gd

(III) was covalently attached to a dendrimer to create a fresh macromolecular contrast MRI agent. The

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

54

authors used 2-(4-isothiocyanatobenzyl)-6-methyl diethylenetriaminepentaacetic acid (1B4M-DTPA) and

Cy5.5 as a bifunctional chelating agent. The PAMAM dendrimers covalently attached to the Gd (III)-

DTPA chelates and units of the near-infrared (NIR) fluorescent dye, Cy5.5, to form a dual-modality MRI-

FI agent [58].

The development of new dendrimers as drug-carriers continues to be essential in cancer therapy.

They can be easily functionalized with a diverse variety of ligands to reach the tumor through the

different body barriers in the body with minimal loss of activity. This results in the selectively targeting

and killing of tumor cells without affecting the normal cells and most importantly, with an actively

controlled release mechanism.

III-4. DENDIMER NANOPARTICLES IN COLORECTAL CANCER

Colorectal cancer (CRC) is the third leading cause of cancer death in the U.S. and additionally the

third widely diagnosed cancer in the world [59].

The chemotherapeutic approach alone has not been found to be very efficient in CRC as the drug

molecules may not reach the target site with an effective concentration and suffer a non-specific

distribution, with only a small fraction of the drug reaching the tumor [60, 61].

These unique "polymeric compounds" can form intelligent species after modification, transporting

drugs into specific areas and at the same time can be used for monitoring the state of organs attacked

by cancer cells, as well as the progress of the curing process [13]. Recent advances in nanotechnology

have rendered it an attractive approach for designing novel clinical solutions for CRC [13].

When tumor cells are shed from primary tumors or metastatic sites of early-stage cancer patients

and enter the bloodstream, they are called circulating tumor cells (CTCs). Detection and capture of

CTCs is of upmost importance because they provide information about the tumor origin. However, they

are very challenging to capture due to their low concentrations in peripheral blood. In order to make

advances in this area, Xie et al. [62] reported an effective approach to specifically bind and capture

colon cancer HT29 cells by using multiple Sialyl Lewis X antibodies (aSlex)-conjugated with PAMAM

dendrimers. Since Sialyl Lewis X is a type II carbohydrate antigen for mediating the CRC metastatic

process, it was expected to capture the antigen positive CTCs in artificial blood samples [62] (Figure III-

5A). Results indicated that the conjugate showed the enhanced capture of HT29 cells in a

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

55

concentration-dependent manner and the maximum capture efficiency of 77.88 % was obtained within

1 hour exposure [62]. This work provided a novel conceptual guidance for the effective prevention of

cancer metastasis.

Capecitabine is one of the most used anticancer drugs in CRC, and is converted into 5-FU by

various metabolic enzymes involved in DNA damage and tumor growth inhibition [63]. However,

capecitabine has multiple adverse side effects affecting the blood, hair cells, bone marrow, and liver. In

this context, Nabavizadeh et al. [61] induced colon adenocarcinoma in mouse models with

azoxymethane, a carcinogen agent, and then investigated the potentiality of non-modified G4 PAMAM

dendrimers to improve capecitabine therapeutic index and decrease its adverse side effects on liver and

bone marrow [61]. Although no targeting was performed, the team compared the effects of free and

conjugated capecitabine form on tumor size and blood cell lines abnormalities. Results showed reduced

side effects in the liver and blood along with decreased tumor size when compared with the free form

[61].

In a recent work, Aliboland et al. [12] encapsulated gold (Au) NPs inside PAMAM dendrimer NPs.

The aim of this study was to investigate the theranostic capability of curcumin-loaded dendrimer-gold

hybrid structures (Figure III-5B). Moreover, the curcumin-loaded PEGylated Au dendrimer was further

conjugated to MUC-1 aptamer (Apt-PEG-AuPAMAM-CUR) in order to target the mucin-1 overexpressing

cancer colorectal adenocarcinoma cells in vitro and in vivo. The obtained results confirmed higher

cellular uptake, internalization and cytotoxicity of Apt-PEG-AuPAMAM-CUR in comparison with PEG-

AuPAMAM-CUR in C26 and HT29 colorectal cancer cell lines. Moreover, the system worked as an

effective anti-tumor therapy and accurate computed tomography imaging of C26 tumor-bearing mice

due to gold accumulation [12].

Similarly, Alibolandi et al. [64] developed camptothecin-loaded pegylated PAMAM dendrimer

(Figure III-5C). For targeting purposes, the team functionalized the system with AS1411 anti-nucleolin

aptamers for site-specific targeting against CRC cells, which overexpresses nucleolin receptors [64].

Remarkably, aptamer AS1411 has now entered clinical trials (phase II/III trials) for acute leukaemia

therapy [65]. Comparative in vitro cytotoxicity experiments demonstrated that the targeted

camptothecin loaded-pegylated dendrimers had higher anti-proliferation activity towards nucleolin-

positive HT29 and C26 colorectal cancer cells than nucleolin-negative CHO cell line. The same system

was tested in vivo on C26 tumor-bearing mice with promising results.

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

56

Another anti-cancer drug currently used in the clinics is irinotecan [66]. However, its utility is

limited by its narrow therapeutic index. Theoretically, a possibility to improve therapeutic index consists

on increasing drug exposure in the diseased tissue, without accumulation in the healthy tissues.

Dendrimer nanotechnology offers this possibility, as shown by England et al. [66]. They modified

generation 5 L-lysine dendrimer with a polyoxazoline as a drug delivery vehicle for improving the

therapeutic index of SN-38, the active metabolite of irinotecan (Figure III-5D). This extensive study

comprised different linker technologies to obtain diverse pharmacokinetic profiles of drug release. Three

conjugates with plasma release half-lives of 2.5 h, 21 h, and 72 h were tested for efficacy and toxicity

using a mouse SW620 xenograft model. The linker with a plasma release half-life of 21 h achieved

sustained SN-38 exposure in blood, above the target concentration. Overall, these extensive studies

allowed to identify a linker, a dose and dosing regimen for SN-38 conjugated to polyoxazoline-modified

dendrimer that maximized efficacy and minimized adverse side effects [66].

Narmani et al. [67] studied the anti-cancer efficacy of oxaliplatin (OX) using a nanocarrier system

with enhanced targeting efficacy towards folic acid receptors (FAR) expressing CRC cells in vitro [67].

This system consisted of PAMAM dendrimers G4 imprinted with polyethyleneglycol (PEG) and folic acid.

Polyethelen glycation of polymeric NPs is frequently applied to increase stability, and thus the half-life,

as well as the non-immunogenic and non-antigenic properties [24, 68]. The PEG-PAMAM nano-complex

containing OX was shown to have a superior cellular uptake in SW480 cell line. The cell viability tests

clearly demonstrated the cancer cell growth inhibition effects of PEG-PAMAM-FA-OX. These fundamental

scientific advances (see Table III-2), coupled with practical methods to covalently conjugate a wide

range of bioactive molecules to the surface of a dendrimer or encapsulate them as guest molecules

within void spaces, provide a highly versatile and potentially extremely powerful technological platform

to fight colorectal cancer.

Table III-2 - Summary of different strategies using dendrimer NP for CRC therapy.

Type of Dendrimer Target Anti-cancer drug Reference PAMAM G4 N/A Capecitabine [61]

Gold NPs inside PAMAM

MUC-1 aptamer

Curcumin [12]

Pegylated PAMAM AS1411 anti-nucleolin aptamers

Camptothecin [64]

L-lysine dendrimer G5 modified with

N/A SN-38 (active metabolite of Irinotecan)

[66]

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

57

polyoxazoline PAMAM G4 Folic acid Oxaliplatin [67]

Figure III-4 - Different strategies for employing dendrimer NPs in CRC research. A) Two synthetic procedures of aSlex-

conjugated dendrimers with FITC labeling by sequential conjugation of the completely carboxylated PAMAM

dendrimers with FITC and aSlex (a–c–e) or by sequential conjugation of the partially carboxylated PAMAM

dendrimers with aSlex and IgG/IgM-FITC (b–d–f). 1, PAMAM with amine groups; 2, PAMAM with the partially

carboxylated groups; 3, PAMAM with the completely carboxylated groups; 4, PAMAM linked with FITC; 5,

PAMAM conjugated with primary antibody; 6, FITC-labeled PAMAM conjugated with antibody; 7, Primary

antibody-coated PAMAM conjugated with FITC-labeled secondary antibody. Reprinted with permission from

[62]. Copyright © 2015 Nature. B) Dendrimer-gold hybrid structure synthesized by complexing AuCl4− ions

with PEGylated amine-terminated generation 5 PAMAM dendrimer. The resultant hybrid system was loaded

with curcumin. The curcumin-loaded PEGylated Au dendrimer was further conjugated to MUC-1 aptamer in

order to target the colorectal adenocarcinoma in vitro and in vivo. Reprinted with permission from [12].

Copyright © 2018 Elsevier. C) Schematic representation of (A) synthesis of pegylated PAMAMdendrimer

(PEG-PAMAM); (B) camptothecin (CPT) loading in the cavities of PEG-PAMAM; (C) conjugation of thiolated

AS1411 aptamers to the maleimide groups of MAL-PEG-PAMAM-CPT and preparation of Apt-PEG-PAMAM-

CPT. Reprinted with permission from [64]. Copyright © 2017 Elsevier. D) Synthetic route to polyoxazoline-

modified dendrimers with linker-SN38 conjugation. Reprinted with permission form [66]. Copyright © 2017

Elsevier.

III-5. CONCLUSIONS

The inefficacy of conventional chemotherapeutic methods has led to the development of new

strategies, which can be used to improve the efficiency of anti-cancer drug delivery into tumors while

minimizing distribution and toxicity in healthy tissues as well as novel imaging tools. These novel

strategies based on the use of NPs loaded with drugs offer unprecedented opportunities both at the

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

58

preclinical and clinical levels. However, some challenges still remain, such as improving the localization,

biodistribution, biocompatibility, and efficacy of these nano-drug systems in vivo, to meet the

requirements of precision cancer diagnosis and therapy. In the current scenario, among the new

nanotechnology platforms, dendrimer-based chemotherapeutics have emerged as one of the most

promising nanotools over the available conventional chemotherapies for the treatment of a variety of

tumors. However, their application in colorectal cancer is still in its infancy.

An ideal therapeutic must have the ability to target cancer cells, image the extent of the tumor and

sense its signatures, deliver a therapeutic, and monitor cells for their response. Although we are not

there yet, this is the goal for our nanotherapeutics.

III-6. REFERENCES

1. Kotelevets, L., et al., Nanotechnologies for the treatment of colon cancer: From old drugs to new hope. International Journal of Pharmaceutics, 2016. 514(1): p. 24-40.

2. Gatoo, M.A., et al., Physicochemical properties of nanomaterials: implication in associated toxic manifestations. Biomed Res Int, 2014. 2014: p. 498420.

3. Tekchandani, P., B.D. Kurmi, and S.R. Paliwal, Nanomedicine to Deal With Cancer Cell Biology in Multi-Drug Resistance. Mini Rev Med Chem, 2017. 17(18): p. 1793-1810.

4. Bar-Zeev, M., Y.D. Livney, and Y.G. Assaraf, Targeted nanomedicine for cancer therapeutics: Towards precision medicine overcoming drug resistance. Drug Resist Updat, 2017. 31: p. 15-30.

5. Maeda, H., H. Nakamura, and J. Fang, The EPR effect for macromolecular drug delivery to solid tumors: Improvement of tumor uptake, lowering of systemic toxicity, and distinct tumor imaging in vivo. Adv Drug Deliv Rev, 2013. 65(1): p. 71-9.

6. Da Silva, C.G., et al., The potential of multi-compound nanoparticles to bypass drug resistance in cancer. Cancer chemotherapy and pharmacology, 2017. 80(5): p. 881-894.

7. Zhitomirsky, B. and Y.G. Assaraf, Lysosomes as mediators of drug resistance in cancer. Drug Resist Updat, 2016. 24: p. 23-33.

8. Villegas, M.R., et al., Multifunctional Protocells for Enhanced Penetration in 3D Extracellular Tumoral Matrices. Chemistry of Materials, 2018. 30(1): p. 112-120.

9. Van Cutsem, E., et al., Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer. N Engl J Med, 2009. 360(14): p. 1408-17.

10. Kopansky, E., Y. Shamay, and A. David, Peptide-directed HPMA copolymer-doxorubicin conjugates as targeted therapeutics for colorectal cancer. J Drug Target, 2011. 19(10): p. 933-43.

11. Landmark, K.J., et al., Synthesis, characterization, and in vitro testing of superparamagnetic iron oxide nanoparticles targeted using folic Acid-conjugated dendrimers. ACS Nano, 2008. 2(4): p. 773-83.

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

59

12. Alibolandi, M., et al., Curcumin-entrapped MUC-1 aptamer targeted dendrimer-gold hybrid nanostructure as a theranostic system for colon adenocarcinoma. International Journal of Pharmaceutics, 2018. 549(1): p. 67-75.

13. Pavitra, E., et al., Engineered nanoparticles for imaging and drug delivery in colorectal cancer. Seminars in Cancer Biology, 2019.

14. Belhadj-Tahar, H., et al., In situ anti-cancer agent derived from [188re]rhenium nitro-imidazole ligand loaded poly-l-lysine dendrimer administrated by direct CT guided stereotactic intrahepatic injection. Journal of Clinical Oncology, 2018. 36(15_suppl): p. e15569-e15569.

15. Araújo, R.V.d., et al., New Advances in General Biomedical Applications of PAMAM Dendrimers. Molecules (Basel, Switzerland), 2018. 23(11): p. 2849.

16. Lee, S., et al., Polyamidoamine (PAMAM) Dendrimers Modified with Cathepsin-B Cleavable Oligopeptides for Enhanced Gene Delivery. Polymers, 2017. 9(6): p. 224.

17. Tomalia, D.A. and J.M.J. Fréchet, Discovery of dendrimers and dendritic polymers: A brief historical perspective*. Journal of Polymer Science Part A: Polymer Chemistry, 2002. 40(16): p. 2719-2728.

18. Oliveira, J.M., et al., Dendrimers and derivatives as a potential therapeutic tool in regenerative medicine strategies—A review. Progress in Polymer Science, 2010. 35(9): p. 1163-1194.

19. Srinageshwar, B., et al., PAMAM Dendrimers Cross the Blood-Brain Barrier When Administered through the Carotid Artery in C57BL/6J Mice. International journal of molecular sciences, 2017. 18(3): p. 628.

20. Hill, E., et al., Synthetic PAMAM-RGD conjugates target and bind to odontoblast-like MDPC 23 cells and the predentin in tooth organ cultures. Bioconjug Chem, 2007. 18(6): p. 1756-62.

21. Salimi, M., et al., Biodistribution, pharmacokinetics, and toxicity of dendrimer-coated iron oxide nanoparticles in BALB/c mice. Int J Nanomedicine, 2018. 13: p. 1483-1493.

22. Thomas, T.P., et al., In vitro targeting of synthesized antibody-conjugated dendrimer nanoparticles. Biomacromolecules, 2004. 5(6): p. 2269-74.

23. Sharma, A.K., et al., Dendrimer nanoarchitectures for cancer diagnosis and anticancer drug delivery. Drug Discov Today, 2017. 22(2): p. 314-326.

24. Castro, R.I., O. Forero-Doria, and L. Guzman, Perspectives of Dendrimer-based Nanoparticles in Cancer Therapy. An Acad Bras Cienc, 2018. 90(2 suppl 1): p. 2331-2346.

25. Hawker, C.J. and J.M.J. Frechet, Preparation of polymers with controlled molecular architecture. A new convergent approach to dendritic macromolecules. Journal of the American Chemical Society, 1990. 112(21): p. 7638-7647.

26. Abbasi, E., et al., Dendrimers: synthesis, applications, and properties. Nanoscale research letters, 2014. 9(1): p. 247-247.

27. Tomalia, D.A., Birth of a new macromolecular architecture: dendrimers as quantized building blocks for nanoscale synthetic polymer chemistry. Progress in Polymer Science, 2005. 30(3): p. 294-324.

28. Palmerston Mendes, L., J. Pan, and V.P. Torchilin, Dendrimers as Nanocarriers for Nucleic Acid and Drug Delivery in Cancer Therapy. Molecules (Basel, Switzerland), 2017. 22(9): p. 1401.

29. Pedziwiatr-Werbicka, E., et al., Dendrimers and hyperbranched structures for biomedical applications. European Polymer Journal, 2019. 119: p. 61-73.

30. Shukla, S.K., P.P. Govender, and A. Tiwari, Chapter Six - Polymeric Micellar Structures for Biosensor Technology, in Advances in Biomembranes and Lipid Self-Assembly, A. Iglič, C.V. Kulkarni, and M. Rappolt, Editors. 2016, Academic Press. p. 143-161.

31. Kaur, D., et al., A review on comparative study of PPI and PAMAM dendrimers. Journal of Nanoparticle Research, 2016. 18(6): p. 146.

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

60

32. Madaan, K., et al., Dendrimers in drug delivery and targeting: Drug-dendrimer interactions and toxicity issues. Journal of pharmacy & bioallied sciences, 2014. 6(3): p. 139-150.

33. Kesharwani, P., K. Jain, and N.K. Jain, Dendrimer as nanocarrier for drug delivery. Progress in Polymer Science, 2014. 39(2): p. 268-307.

34. Chang, H., et al., Surface-engineered dendrimers with a diaminododecane core achieve efficient gene transfection and low cytotoxicity. Bioconjug Chem, 2014. 25(2): p. 342-50.

35. Chauhan, A.S. and M. Kaul, Engineering of “critical nanoscale design parameters” (CNDPs) in PAMAM dendrimer nanoparticles for drug delivery applications. Journal of Nanoparticle Research, 2018. 20(9): p. 226.

36. Oliveira, J.M., et al., Surface Engineered Carboxymethylchitosan/Poly(amidoamine) Dendrimer Nanoparticles for Intracellular Targeting. Advanced Functional Materials, 2008. 18(12): p. 1840-1853.

37. Carvalho, M.R., et al., Colorectal tumor-on-a-chip system: A 3D tool for precision onco-nanomedicine. Sci Adv, 2019. 5(5): p. eaaw1317.

38. Oliveira, J.M., et al., The osteogenic differentiation of rat bone marrow stromal cells cultured with dexamethasone-loaded carboxymethylchitosan/poly(amidoamine) dendrimer nanoparticles. Biomaterials, 2009. 30(5): p. 804-13.

39. Qu, Y., et al., Self-Assembled DNA Dendrimer Nanoparticle for Efficient Delivery of Immunostimulatory CpG Motifs. ACS Applied Materials & Interfaces, 2017. 9(24): p. 20324-20329.

40. Zhao, L., X. Shi, and J. Zhao, Dendrimer-based contrast agents for PET imaging. Drug Delivery, 2017. 24(2): p. 81-93.

41. Huang, Y., et al., Dendrimer-Based Responsive MRI Contrast Agents (G1–G4) for Biosensor Imaging of Redundant Deviation in Shifts (BIRDS). Bioconjugate Chemistry, 2015. 26(12): p. 2315-2323.

42. Caminade, A.-M., et al., The key role of the scaffold on the efficiency of dendrimer nanodrugs. Nature Communications, 2015. 6: p. 7722.

43. Baker, J.R., Jr., Dendrimer-based nanoparticles for cancer therapy. Hematology Am Soc Hematol Educ Program, 2009: p. 708-19.

44. Saluja, V., et al., Smart dendrimers: Synergizing the targeting of anticancer bioactives. Journal of Drug Delivery Science and Technology, 2019. 52: p. 15-26.

45. Mura, S., J. Nicolas, and P. Couvreur, Stimuli-responsive nanocarriers for drug delivery. Nature Materials, 2013. 12: p. 991.

46. Gai, S., et al., Recent advances in functional nanomaterials for light–triggered cancer therapy. Nano Today, 2018. 19: p. 146-187.

47. Choi, S.K., et al., Light-controlled release of caged doxorubicin from folate receptor-targeting PAMAM dendrimer nanoconjugate. Chem Commun (Camb), 2010. 46(15): p. 2632-4.

48. Nigam, S. and D. Bahadur, Dendrimer-conjugated iron oxide nanoparticles as stimuli-responsive drug carriers for thermally-activated chemotherapy of cancer. Colloids and Surfaces B: Biointerfaces, 2017. 155: p. 182-192.

49. Amreddy, N., et al., Chemo-biologic combinatorial drug delivery using folate receptor-targeted dendrimer nanoparticles for lung cancer treatment. Nanomedicine: Nanotechnology, Biology and Medicine, 2018. 14(2): p. 373-384.

50. Chittasupho, C., S. Anuchapreeda, and N. Sarisuta, CXCR4 targeted dendrimer for anti-cancer drug delivery and breast cancer cell migration inhibition. European Journal of Pharmaceutics and Biopharmaceutics, 2017. 119: p. 310-321.

Chapter III – Dendrimer Nanoparticles in Colorectal Cancer Research

61

51. Yousef, S., et al., Development of asialoglycoprotein receptor directed nanoparticles for selective delivery of curcumin derivative to hepatocellular carcinoma. Heliyon, 2018. 4(12): p. e01071-e01071.

52. Lee, T.J., et al., RNA Nanoparticle-Based Targeted Therapy for Glioblastoma through Inhibition of Oncogenic miR-21. Mol Ther, 2017. 25(7): p. 1544-1555.

53. Mishra, V. and P. Kesharwani, Dendrimer technologies for brain tumor. Drug Discovery Today, 2016. 21(5): p. 766-778.

54. Liu, S., et al., Gene and doxorubicin co-delivery system for targeting therapy of glioma. Biomaterials, 2012. 33(19): p. 4907-16.

55. Wu, X.X., et al., Doxorubicin enhances TRAIL-induced apoptosis in prostate cancer. Int J Oncol, 2002. 20(5): p. 949-54.

56. Sharma, A.K., et al., Dendrimer nanoarchitectures for cancer diagnosis and anticancer drug delivery. Drug Discovery Today, 2017. 22(2): p. 314-326.

57. Zhou, Z. and Z.-R. Lu, Gadolinium-based contrast agents for magnetic resonance cancer imaging. Wiley interdisciplinary reviews. Nanomedicine and nanobiotechnology, 2013. 5(1): p. 1-18.

58. Talanov, V.S., et al., Dendrimer-based nanoprobe for dual modality magnetic resonance and fluorescence imaging. Nano Lett, 2006. 6(7): p. 1459-63.

59. Siegel, R.L., K.D. Miller, and A. Jemal, Cancer statistics, 2018. 2018. 68(1): p. 7-30. 60. Yan, W., et al., Overcoming Drug Resistance in Colon Cancer by Aptamer-Mediated Targeted

Co-Delivery of Drug and siRNA Using Grapefruit-Derived Nanovectors. Cellular Physiology and Biochemistry, 2018. 50(1): p. 79-91.

61. Nabavizadeh, F., et al., Evaluation of Nanocarrier Targeted Drug Delivery of Capecitabine-PAMAM Dendrimer Complex in a Mice Colorectal Cancer Model. Acta Med Iran, 2016. 54(8): p. 485-493.

62. Xie, J., et al., Multivalent Conjugation of Antibody to Dendrimers for the Enhanced Capture and Regulation on Colon Cancer Cells. Scientific Reports, 2015. 5: p. 9445.

63. Gulbake, A., et al., Insight to drug delivery aspects for colorectal cancer. World journal of gastroenterology, 2016. 22(2): p. 582-599.

64. Alibolandi, M., et al., Smart AS1411-aptamer conjugated pegylated PAMAM dendrimer for the superior delivery of camptothecin to colon adenocarcinoma in vitro and in vivo. Int J Pharm, 2017. 519(1-2): p. 352-364.

65. Fan, X., et al., Bioactivity of 2'-deoxyinosine-incorporated aptamer AS1411. Sci Rep, 2016. 6: p. 25799.

66. England, R.M., et al., Tumour regression and improved gastrointestinal tolerability from controlled release of SN-38 from novel polyoxazoline-modified dendrimers. J Control Release, 2017. 247: p. 73-85.

67. Narmani, A., et al., Targeting delivery of oxaliplatin with smart PEG-modified PAMAM G4 to colorectal cell line: In vitro studies. Process Biochemistry, 2018. 69: p. 178-187.

68. Fox, M.E., et al., Synthesis and in vivo antitumor efficacy of PEGylated poly(l-lysine) dendrimer-camptothecin conjugates. Molecular pharmaceutics, 2009. 6(5): p. 1562-1572.

62

SECTION 2

MATERIALS AND

METHODS

Chapter IV – Materials and Methods

63

Chapter IV

Materials and methods

Chapter IV – Materials and Methods

64

Chapter IV

Chapter IV - Materials and Methods

OVERVIEW

This chapter intends to provide a detailed overview on the experimental procedures behind

the results presented in Section 3. Moreover, some considerations will be made regarding the

selection of the materials and hydrogels processing methodologies, the microfluidic devices, as

well as the physicochemical and biological characterization techniques used to sustain the

proposed methodologies. With this, it is expected to present a clearer perspective of the

developed work in this thesis and how it can be correlated. Despite all the following chapters

already having a “materials and methods” sub-section, herein these are presented in a general

perspective, while providing more detailed information in all the specific steps followed for each

work.

Chapter IV – Materials and Methods

65

IV-1. MATERIALS

European Society for Biomaterials’ (ESB), current definition for “biomaterial” is a material

intended to interface with biological systems to evaluate, treat, augment or replace any tissue,

organ or function of the body [1]. Natural biomaterials were considered vital in the conception of

this thesis. The TE systems developed in this thesis were based on different types of

biomaterials: Carboxymethylchitosan to modify the dendrimer nanoparticles, silk and Matrigel as

extra-cellular matrix mimics in the developed 3D tumor models and silk fibroin as the biomaterial

to produce a microfluidic platform.

IV-1.1. Carboxymethylchitosan (CMCht)

Chitin and chitosan are attractive macromolecules, but their use is restricted due to their

poor water-solubility (Figure IV–1) [2]. Chemical modifications in chitosan molecular structure are

feasible due to the reactive functional groups (e.g. amine -NH2 and hydroxyl -OH), allowing to

tailor its degradation profile and widen its applications [3]. Thus, to develop biomaterials with

improved physicochemical properties, attempts have been made to modify its structure by

introducing acetyl, carboxymethyl, sulfuryl, or phosphoryl groups, among others [4]. In one hand,

chitin is a linear, high molecular weight, crystalline polysaccharide, theoretically comprised

entirely by N-acetylated D-glucosamine units. On the other hand, chitosan, known as poly-(β-

1/4)-2-amino-2-deoxy-D-glucopyranose is a hydrophilic polycationic copolymer made of D-

glucosamine and N-acetyl-D-glucosamine, which are linked with β (1–4) bonds.

Chapter IV – Materials and Methods

66

Figure IV-1 - Chemical structure of chitin and chitosan. Reprinted with permission from [5].

Deacetylation of chitin results in chitosan, a polymer, widely studied for its pharmaceutical

and non-pharmaceutical applications. The hurdle in comprehending these applications is its

limited solubility. Carboxymethylation of chitosan helps to surmount this hurdle with its improved

solubility in water. Carboxymethylchitosan (CMCht) is soluble in a wide range of pH and its

properties, namely low toxicity and antimicrobial activity, may wider its application, including

medical and pharmaceutical areas - mainly for the controlled release of drugs [6].

The chitosan derivatives obtained by its carboxymethylation are old in the art with three

types: N-carboxymethyl chitosan (NCMC) [7], O-carboxymethyl chitosan (OCMC) [8], and N,O-

carboxymethyl chitosan (NOCMC) [9], being the choice of the appropriate reaction conditions and

reagents determinant for the preparation of N-, O-, N,O- or N,N-dicarboxymethylchitosan (Figure

IV-2).

Chapter IV – Materials and Methods

67

Figure IV-2 - Preparative methods of different types of Carboxymethylchitosan. Reprinted with permission from [10].

Thus, the properties and applications of carboxymethylchitosan are strongly dependent on

its structural characteristics, mainly the degree of substitution (DS). For the purpose of this

thesis, reagent grade chitosan particles with a DS of ~91 %, monochloroacetic acid, 1 % acetic

acid solution, 40 % sodium hydroxide solution, and acetone were used in the preparation of

CMCht. The methodology used has been previously reported by Chen et al. [11]. Briefly, chitosan

(5 g), 40 % sodium hydroxide solution and isopropanol (100 mL) were added into a flask (500

mL) to alkalize at RT for 1 hour. Then, monochloroacetic acid (18 g) was added into the reaction

vessel dropwise for 30 minutes and reacted for 2 hours at 60 ºC. The reaction was stopped by

adding 70 % ethanol (200 mL). The solid was filtered and rinsed in 70-90 % ethanol to de-salt

and de-water, and vacuum dried at room temperature. Finally, the H-form of CMCht was obtained

by suspending the Na salt CMCht in 80 % ethanol aqueous solution (100 mL) and 37 %

hydrochloric acid (10 mL) for 30 minutes, under agitation. The solid was filtered and rinsed in

70-90 % ethanol to neutral, frozen at -80 ºC and freeze-dried (Telstar-Cryodos -80, Spain). CMCht

with a DD of 80 % and DS of 47 % was chosen for being used in the production of composite

scaffolds and to surface modify the dendrimers, due to the previously highlighted advantages.

Chapter IV – Materials and Methods

68

IV-1.2. Silk fibroin

Silk fibers are obtained from various animals such as spiders, silkworms, scorpions, mites

and flies [12]. Silk is widely considered a good source for the development of biomedical devices.

Being one of the oldest natural polymers discovered by men, silk has evolved during its history of

over 380 million years [13]. It possesses good biocompatibility, suitable mechanical properties

and is produced in bulk in the textile sector [14]. The unique combination of elasticity and

strength along with mammalian cell compatibility makes silk fibroin an attractive material for

tissue engineering. These fibers are composed of a fibrous protein core, composed of fibroin,

with a surrounding glue protein, sericin. More specifically, two parallel fibroin fibers are held

together with a layer of sericin on their surfaces. According to the silk fibers origin, its amino acid

composition varies, as can be seen in Table IV-1.

Table IV-1 - Comparative data on amino acid composition of B. mori, N. edulis and N. clavipes silks. Reprinted with

permission from [15].

Amino acid B. mori N. edulis N. clavipes

Ala 30.0 24.0 22.71

Gly 42.9 38.2 9.96

Tyr 4.8 5.2 2.99

Ser 12.2 1.7 2.24

Asp 1.9 1.5 1.06

Arg 0.5 2.3 1.76

His 0.2 Trace 0.21

Glu 1.4 12.8 11.02

Lys 0.4 0.3 0.10

Val 2.5 0.7 0.89

Leu 0.6 2.4 4.26

Ile 0.6 0.4 0.07

Phe 0.7 0.3 0.26

Pro 0.5 9.4 2.04

Thr 0.9 0.4 0.34

Met 0.1 0.3 0.04

Chapter IV – Materials and Methods

69

Cys 0.0 ND 0.06

Silkworm's silk is an established fiber extensively used. On the other hand, the cannibalistic

nature of spiders restricts its commercial production of silk [16]. Additionally, the yield of fiber

from a single silk cocoon is 600–1500 m, compared to only ~137 m from a spider [16]. Spider

silks are also heterogeneous in nature. Therefore, silk-based biomaterials are commonly

prepared from silkworm silk. In this thesis, the silk used was produced by Bombyx mori, a

member of the Bombycidae family. B. mori silk is also known as mulberry silk. Another silk

producing family is Saturniidae and the silk is known as non-mulberry silk.

Depending on the silkworm sources, different processing methodologies are used for silk

purification. In the case of the Bombyx mori silk fibroin, this process is routinely established,

involving a simple alkali- or enzyme-mediated degumming procedure to extract sericin from the

cocoons and obtain SF at 70 % off the original cocoon mass [17, 18].

Upon degumming the raw silk to remove the sericin, the obtained fibroin fibers appear

shiny and feel soft to the touch [19, 20]. Moreover, the fibroin fibers are endowed with a

combination of attractive strength, toughness, biocompatibility/biodegradability and thermal

stability, representing one of the most impressive natural protein fibers [21].

Sericin (20-310 kDa) is a gum-like protein produced with silk, currently suggested by the

scientific community as an inducer of in vivo inflammatory response in its native state [22],

vindicating the necessity to extract this protein from the cocoons before processing silk fibroin

(SF) for biomedical applications.

However, recent studies now claim that silk sericin can accelerate cell proliferation and

attachment; however, sircerin can be extracted by various methods, which results in different

physical and biological properties. Also, through different treatments, SF can be prepared to hold

a wide range of forms, such as solution, powder, fibers, films, hydrogels, and sponges; this

allows the use of SF for constructing many different scale structures, ranging across the nano-,

micro-, and macro-dimensions [23]. Related to its chemical structure, SF is formed by heavy (H)

and light (L) chains linked by a disulfide bond, with hydrophobic domains of the H-chains

interspaced by hydrophilic regions at the L-chains. The hydrophobic domains contain highly

ordered Glycine-X (X being Alanine, Serine, Tyrosine, Valine) repeats that form stacked anti-

Chapter IV – Materials and Methods

70

parallel β-sheets (silk II) of crystalline structure. The hydrophilic regions constitute the amorphous

phase of the protein. Processed in a variety of forms, silk-based materials include porous

scaffolds [24] and sponges [25], films [26], tubes [27], electrospun [28] or micro-fibers [29],

hydrogels [30] and textiles [31].

There are several developed protocols and techniques to transform SF raw material into

regenerated SF solutions, usually involving harsh, chaotic solvents such as strong acids or ionic

concentrated salt solutions, such as the organic solvent hexafluoroisopropanol (HFIP),

concentrated lithium bromide solution, ionic liquids, lithium thiocyanate, or a calcium

chloride/ethanol/water ternary solvent [32-34]. The constructs produced from these solutions

acquire β-sheet conformation. Temperature and pH variations, ultrasonication or vortex also

induce β-sheet conformation. The presence of β-sheet crystalline structure in SF has shown to

significantly improve the mechanical properties, water insolubility and degradability of the SF-

based structures. However, SF hydrogels and membranes prepared in an amorphous state, also

showed good specificities for soft tissues clinical application materials to be prepared in a friendly

processing, which favors the incorporation of bioactive proteins, drugs and cells [35].

. Thus, SF has been exploited with different biomedical purposes, as a result of its

processing versatility and extensive knowledge on how to control its properties, including

mechanical strength, elasticity, biocompatibility, and controlled biodegradability. Therefore, in this

thesis, Bombyx mori silk cocoons were used purified in order to isolate SF used for preparing SF-

based hydrogels, either as an alternative to Matrigel as ECM material in Chapter VI, or for the

production of a microfluidic platform in Chapter IX. For the production of silk in this thesis, all the

cocoons were supplied by the Portuguese Association of Parents and Friends of Mentally

Disabled Citizens (APPACDM, Castelo Branco, Portugal).

IV-1.3. Matrigel®

Matrigel® is a soluble and sterile extract of basement membrane proteins derived from the

Engelbreth-Holm-Swarm (EHS) tumor, crosslinking into a 3D gel at 37 ºC, able to support cell

morphogenesis, differentiation, and tumor growth (Figure IV-2) [36].

Matrigel® was first subject to scientific publication in 1977, developed by Roslyn Orkin who

studied a mouse tumor with an abundant extracellular matrix, originally identified as a poorly

Chapter IV – Materials and Methods

71

differentiated chondrosarcoma [37]. Structural and amino acid analyses proposed that the tumor

matrix was distinct from cartilage, and instead resembled basement membrane. It was finally

named EHS tumor to acknowledge J. Engelbreth-Holm of Denmark, who discovered it, and

Richard Swarm, who maintained and characterized it [38].

Methods for culturing mammalian cells ex vivo are increasingly needed to study cell and

tissue physiology and to grow replacement tissue for regenerative medicine. Two-dimensional

culture has been the paradigm for typical in vitro cell culture, however, it has been demonstrated

that cells behave more natively when cultured in three-dimensional environments [39].

The in vivo and in vitro extracellular matrix (ECM) is known to play a key role in numerous

outcomes that direct cell fate and behavior [40, 41]. Characteristic ECM proteins comprise

laminin, collagens, glycoproteins, and proteoglycans [42]. The main function of the ECM is to

support the growth and maintenance of a variety of cells. In vitro growth matrices can be a

variety of materials, such as chemically treated culture dish plastic, or layers of deposited

protein. Often these matrices are very simple, consisting of a mixture of purified proteins such as

collagen and laminin. Matrices such as these, which contain only major ECM proteins, are not

applicable to all cell types [43].

The basement membrane extracellular matrix contacts epithelial, endothelial, fat and

smooth muscle cells. Because this extracellular matrix is so thin, it has been hard to study its

composition, structure, and function. The extract of a tumor was found to contain all of the

components present in basement membrane and to be very biologically active. This extract, sold

as Matrigel®, Cultrex®, or EHS matrix®, acts as a substrate or physical support for cultured

cells helping to create more in vivo-like extracellular matrices promotes cell differentiation, and is

used to measure the invasive activity of tumor cells. In vivo, it is used for migration assays [44],

sprouting assays [45], capillary-like tube formation [46], differentiation of stem cells [47],

measuring angiogenic inhibitors and stimulators [48], to improve graft survival [49], repair

damaged tissues, and increase tumor growth [39, 50] and tumor cell interactions study [51].

However, Matrigel® is not a well-defined matrix, and therefore can produce a source of

variability in experimental results. The presence of several growth factors was detected (basic

fibroblast growth factor, epidermal growth factor, insulin‐like growth factor 1, transforming growth

Chapter IV – Materials and Methods

72

factor β, platelet‐derived growth factor, and nerve growth factor) in standard Matrigel® through

the use of immunoassays.

Figure IV-3 – Matrigel® in its original packing and after cell encapsulation.

IV-1.4. Polydimethylsiloxane (PDMS)

Polydimethylsiloxane (PDMS) [(CH3)2SiO]n is by far the most widely used polymer in siloxane

(“silicone”) elastomers [52]. Substituent groups other than methyl, such as phenyl or

trifluoropropyl can be incorporated, often as copolymers with PDMS. Apart from the effects of

cross-links, the macromolecular chains must be free to move reversibly past one another. PDMS

is a mineral-organic polymer (containing carbon and silicon) of the siloxane family (word derived

from silicon, oxygen and alkane). It has been used as a food additive (E900) in shampoos, and

as an anti-foaming agent in beverages [53].

In the last years, rapid progress in micro and nanotechnologies such as lab-on-a-chip

(microfluidic networks, sensors, actuators, and connectors), soft lithography (replica moulding,

microcontact printing and affinity contact printing), and stretchable transparent electronics has

strongly benefitted from high-performance polymers like PDMS that are suited for high-fidelity

microsystem construction and rapid prototyping [54]. For the fabrication of microfluidic devices,

PDMS (liquid) mixed with a cross-linking agent is poured into a microstructured mold and heated

to obtain an elastomeric replica of the mold. The PDMS is also easy to mold, because, even

when mixed with the cross-linking agent, it remains liquid at RT for many hours. The PDMS can

mold structures at high resolutions inclusive of a few nanometers [55].

Chapter IV – Materials and Methods

73

Furthermore, PDMS oxidation using plasma changes the PDMS surface chemistry and

produces silanol terminations (SiOH) on its surface [56]. This process also makes the surface

resistant to the adsorption of hydrophobic and negatively-charged molecules. In addition, PDMS

plasma oxidation is used to functionalize the PDMS surface with trichlorosilane or to covalently

bond PDMS (at the atomic scale) on an oxidized glass surface by the creation of a Si-O-Si bonds

[55].

PDMS presents several characteristics that allowed it to be the most widely used polymer in

microfluidics:

a) It is transparent at optical frequencies (240 nM – 1100 nM), allowing for content

observation in microchannels, either visually or through a microscope; also, it has a low auto-

fluorescence [57];

b) It is gas permeable and considered “cell friendly” (with few restrictions) [58];

c) The PDMS bonds tightly to glass (Figure IV-4) or another PDMS layer with a simple

plasma treatment, allowing the production of multilayers of PDMS;

d) During cross-linking, it can be coated with a controlled thickness on a substrate using a

simple spin coating. This allows the fabrication of multilayer devices and the integration of micro

valves;

e) It is deformable, which allows the integration of microfluidic valves using the deformation

of PDMS micro-channels, the easy connection of leak-proof fluidic connections and its use to

detect very low forces like biomechanics interactions from cells [59];

f) It is inexpensive compared to previously used materials (e.g. silicon).

Chapter IV – Materials and Methods

74

Figure IV-4 – PDMS before and after crosslinking and plasma treatment (adhered to glass).

IV-1.5. YIGSR Peptide

Laminin-111 is a large trimeric basement membrane glycoprotein with many active sites. In

particular, four peptides active in tumor malignancy studies have been identified in laminin-111

using a systematic peptide screening method. Two of the peptides (IKVAV and AG73) are found

on the α1 chain, one (YIGSR) (Figure IV–5) of the β1 chain and one (C16) on the γ1 chain [60].

Chapter IV – Materials and Methods

75

Figure IV-5 - Schematic model of Laminin-111 showing the location of peptides that exhibit cell attachment activity.

Laminin-111 is composed of three subunits, α, β and ϒ chains. Forty-five active peptides are

localized in α1 chain, 14 active peptides in β1 chain and 12 active peptides in ϒ1 chain. The four

highlighted peptides described here are active in tumor malignancy. Reprinted with permission from

[60].

The first described and most studied Laminin-111-derived active peptide, YIGSR (Tyr–Ile–

Gly–Ser–Arg),(Figure IV-5) from the β1 chain binds to the 32/67 kDa cell surface receptor and

has many activities related to its inhibition of malignancy [60].

The peptide YIGSR comprised of residues 929–933 on the β1 chain has been found to

inhibit tumor growth and metastasis [61]. Literature shows that both polymerized and

polyethylene glycol-conjugated YIGSR peptides significantly improved the inhibitory effect of tumor

metastasis [62]. YIGSR peptide was also shown to stimulate the apoptosis of fibrosarcoma cells,

Chapter IV – Materials and Methods

76

but not of colon adenocarcinoma cells, suggesting cell-type specificity, which is not totally

understood [63].

Regarding its receptor, while 37/67 LR was the first laminin receptor to be identified, its

characterization is still incomplete and has been complicated by the fact that it is also involved in

a variety of other unrelated roles. Indeed, beside its ability to interact directly with laminin

through most likely the YIGSR sequence on the laminin β1 chain short arm, 37/67 LR can play

additional roles in the cell [64].

Various groups have used the YIGSR peptide to localize tumors cells and to target tumor

cells with drug delivery based on the peptide binding to the 32/67 kDa receptor [60]. This is

because YIGSR laminin receptor binding peptide has demonstrated to be a promising class of

targeting ligand for specific drug delivery in cancer treatment. Cancer cells from various sources

have been reported to express high levels of the 67 kDa laminin receptor (67 LR) [64], including

bile duct carcinoma, colorectal carcinoma, cervical cancer, and breast carcinoma [65], which

have shown to correlate with tumor invasiveness [66]. Earlier work have demonstrated that the

laminin binding capacity of plasma membranes isolated from human invasive breast carcinomas

is 50-fold greater than that of membranes from benign lesions or normal tissue [67].

Figure IV-6 Structure of Laminin-related peptide TYR-ILE-GLY-SER-ARG.

Active targeting is usually based on coupling of tumor-specific ligands on the NPs, resulting

in specific attraction between the NPs and the cancerous cells. Cancer cells from various sources

have been reported to express high levels of the 67 kDa laminin receptor, the entity responsible

for the interactions between cells and laminin of the extracellular matrix.

Chapter IV – Materials and Methods

77

In this thesis, specifically in Chapter VIII it is described the development of

carboxymethylchitosan/poly(amidoamine) (CMCht/PAMAM) dendrimer nanoparticles

functionalized with YIGSR laminin receptor binding peptide for the actively targeting and specific

delivery of therapeutic agents into colorectal cancer cells.

IV-2. REAGENTS

Unless addressed otherwise, all the reagents used in this thesis were purchased from Sigma-

Aldrich (St. Louis, MO, USA).

IV-3. METHODOLOGIES FOR PROCESSING OF BIOMATERIALS

Regarding the fabrication of the scaffold itself, there is no idyllic methodology or common

biomaterial processing that meets all the scaffolding requirements for tissue regeneration.

Different tissues, as well as different biomaterials, require specific physical, mechanical and

degradation properties [68].

Knowing that the classic TE paradigm relies on a combination of biomaterial scaffolds, cells,

and bioactive molecules to orchestrate tissue formation and integration within the host

environment [69], biomaterials processing plays a vital role in the regenerative process. TE

strategies are premeditated to influence the physical, chemical and biological environment

surrounding the damaged tissue. The processing method is the key factor that mostly contributes

to control all the parameters that will influence its behavior in vitro and in vivo [70]. The different

biomaterials used in this thesis were processed in a variety of ways, namely in the form of freeze-

dried nanoparticles, membranes or hydrogels.

IV-4. DENDRIMER NANOPARTICLE PRODUCTION

Many scientific efforts are directed at improving the characteristics of biomacromolecules for

applications as host-guest chemistry, nanocarriers, and catalysts, among others. As discussed in

Chapter IV – Materials and Methods

78

Chapter III, dendrimers are among such type of macromolecules since it offers a lot of

possibilities in the molecular design, namely in the field of cancer research.

Dendrimers are well-defined, multivalent molecules, having branched structures of nanometer

size, and also referred to as “star-like polymers” [71, 72]. The two most commonly used

dendrimers are polyamidoamine (PAMAM) dendrimers and polypropylene imine dendrimers [73].

The terminal functional group of dendrimers can easily be utilized for engineering the interaction

between molecules of interest with dendrimers. Such chemical versatility also allows producing

multifunctional nanobiomaterials.

The biotechnological applications of dendritic polymers in nanotechnology and regenerative

medicine are immense, as thoroughly described by Oliveira et al. [74]: from osteochondral

application [75], central nervous system [76], gene delivery [77] and theranostic applications

[78], dendrimers have caught the attention of scientific community in the last decades.

Dendrimers may be created by two main methods: In the divergent method, employed in

early times, the synthesis begins from the core of the dendrimers to which the arms are attached

by adding building blocks in an extensive and step-wise manner. This process enables

dendrimers with incrementally cumulative generation numbers. Nonetheless, only one type of

reaction can be executed at each step, yielding an even exhibition of only one functional group on

the exterior surface. Major advantages are associated with this type of synthesis, namely the

speed of the process, cheap reagents, exponential growth and possibility to prepare large

dendrimers [79]. In the convergent method, initiated by Fréchet et al. [80], synthesis begins from

the exterior, starting with the molecular structure that eventually becomes the outermost arm of

the final dendrimer structure. In this tactic, the ultimate generation number is predetermined,

compelling the synthesis of branches of the different desired sizes in advance for each generation

[80].

In this thesis, we modify amine-terminated PAMAM dendrimers with water-soluble CMCht as

an attempt to functionalize the dendrimers and neutralize the amine groups, which have been

found to be accountable for the toxicity of the dendrimers [81]. Research priorities for

intracellular delivery and imaging are addressed simultaneously.

Chapter IV – Materials and Methods

79

IV-4.1. Carboxymethylchitosan/poly(amidoamine) dendrimer (CMCht/PAMAM) nanoparticles

For the synthesis of the CMCht/PAMAM dendrimer macromolecules, carboxymethylchitosan

with a DD of 80 % and DS of 47 % was used. Starburst® poly(amidoamine)-carboxylic terminated

dendrimers, PAMAM-CT (G1.5, 20 % (w/v) methanolic solution) with an ethylenediamine core

were purchased from Aldrich. CMCht/PAMAM nanoparticles were prepared in a stepwise manner

(divergent method) as follows: (i) increasing the generation of the PAMAM-CT (G1.5), (ii) obtaining

a PAMAM-methyl ester terminated dendrimer, (iii) reaction of PAMAM and CMCht, i.e. the

reaction goes through a condensation reaction between the methyl ester and amine groups, and

(iv) modifying the methyl ester groups that do not react with carboxylic ones in the

CMCht/PAMAM dendrimer, followed by precipitation. The increase of the dendrimers’ generation

was carried by means of transferring an appropriate volume of PAMAM-CT (G1.5) in methanol

(MeOH) to a 2 mL volumetric flask. The solvent was evaporated off under nitrogen gas, and the

traces dried under vacuum in order to completely remove the methanol. The starting compound

was re-dissolved in ultra-pure water to give a final concentration of 10 mg/mL and the pH was

adjusted to 6.5 with dilute hydrochloric acid solution (HCl) solution. Then, 1-ethyl-3(3-

dimethylaminopropyl) carbodiimide hydrochloride (EDC) was added to the solution at a molar

ratio sufficient to modify the carboxylate residue of the dendrimers and kept under agitation for

30 minutes at RT. Ethylenediamine (EDA) was added to the solution at a molar ratio equal to that

of EDC and let react for at least 4 hours. After this period, the exceeding EDC was removed by

dialysis (benzoylated cellulose tubing for separating compounds with a molecular weight of ≤

1.200 from compounds with a molecular weight > 2,000). The compound was used without

purification in the next step. After preparing the poly(amidoamine)-amine terminated (PAMAM-

AT), an exhaustive alkylation of primary amines (Michael addition) was carried [82]. An

appropriate volume of PAMAM-AT (~8.4 mmol) was transferred to a 50 mL flask and 30 mL of

MeOH and 1.14 mL of methyl methacrylate (MA, ~12.6 mmol) were added. The solution was

kept under agitation in a water bath for 24 hours at 50 °C, to obtain the compound (PAMAM-

methyl ester). The carboxymethylchitosan (100 mg) dissolved in ultrapure water (10 mL) was

mixed with the latter obtained PAMAM-methyl ester dendrimer (50 mg), which was previously

dissolved in a 20/80 water/methanol (v/v) solution. The final solution was diluted by adding 30

mL of methanol and kept under agitation for 72 hours. After this period, CMCht/PAMAM

dendrimer with carboxylic-terminated groups were obtained. CMCht/PAMAM dendrimer

Chapter IV – Materials and Methods

80

nanoparticles were then precipitated after addition of an appropriate volume of a saturated

sodium carbonate, Na2CO3 solution and acetone.

The synthesis of CMCht/PAMAM was crucial in this thesis. In Chapter V, the

CMCht/PAMAM effect on a variety of cancer cell lines’ viability and internalization levels under

static and dynamic culturing were analyzed. A number of different cancer cell types such as

glioblastoma, cervical cancer and colorectal cancer cells were screened.

IV-4.2. Incorporation of Gemcitabine (GEM) into the carboxymethylchitosan/poly(amidoamine)

dendrimer nanoparticles

CMCht/PAMAM dendrimer nanoparticles were mixed with a GEM solution with a final

concentration of 1x10-4 M under agitation (w/w). The mixture was then added to the precipitation

media consisting of a saturated sodium carbonate, (Na2CO3) and acetone solution, under vigorous

agitation. Precipitates were collected by filtration and dispersed in ultrapure water for dialysis

during the period of 48 hours. Both CMCht/PAMAM and GEM-loaded CMCht/PAMAM dendrimer

nanoparticles were obtained by freezing the solution at -80 °C and freeze-drying (Telstar-Cryodos

-80, Spain) up to 7 days to complete remove the solvent. It is noteworthy that the

CMCht/PAMAM dendrimer nanoparticles and GEM-loaded CMCht/PAMAM dendrimer

nanoparticles are water-soluble at physiological pH.

IV-4.3. Investigation of GEM release from CMCht/PAMAM dendrimer nanoparticles

Encapsulation efficiency, drug loading efficiency and release profile studies were performed

for the use of anti-cancer drug GEM. For assessing encapsulation efficiency, CMCht/PAMAM (50

mg) was dissolved in 50 mL deionized water. GEM, (7 mg) in 2 mL deionized water was added

drop-wise to the dendrimer solution and stirred for 30 minutes. Precipitation of the dendrimers

and consequent encapsulation of the drug was induced as described before. Samples were taken

before and after precipitation, centrifuged, analyzed by UV-Vis spectrophotometer at 275 nm, in a

Hellma Quartz Plate, and compared to a calibration curve.

For evaluating drug-loading efficiency, pre-weighted amount of GEM-loaded CMCht/PAMAM

(50 mg) was added to distilled water and sonicated to extract GEM. The extracted GEM

Chapter IV – Materials and Methods

81

concentration was analyzed by UV-Vis spectrophotometer at 270 nm and compared to a

calibration curve. The absorbance was used to calculate the percentage of GEM encapsulated.

The following equations (Eq.) were used:

Equation IV-1 – Encapsulation efficiency.

Equation IV-2 – Drug loading efficiency.

GEM release from nanoparticles was measured by UV spectrophotometry (Shimadzu) at 275

nm, in a Hellma Quartz Plate, after dissolution of 50 m of GEM-loaded CMCht/PAMAM

dendrimer nanoparticles in 5 mL of phosphate buffer saline solution (PBS) with sodium azide

0.01 % (w/v). The in vitro release was performed at 37 ºC under stirring at 60 rpm.

These studies were performed on Chapter VII, when GEM was encapsulated in the

dendrimer nanoparticles and applied and released in a dynamic microfluidic platform, which

aimed at mimicking the 3D colorectal tumor microenvironment.

IV-4.4. Labeling of carboxymethylchitosan/poly(amidoamine) dendrimer nanoparticles with

fluorescein isothiocyanate

Labeling the dendrimer NPs with a fluorescent molecule is of crucial importance, either as a

tracking tool to assess their uptake, and to track the delivery of encapsulated drugs in real time.

For this, fluorescein isothiocyanate (FITC) was chosen.

In order to label the CMCht/PAMAM dendrimer with FITC, a 10 mg/mL FITC solution was

prepared in anhydrous dimethyl sulfoxide (DMSO), in dark conditions. Conjugates of FITC-

CMCht/PAMAM were prepared by covalently bonding the amine group of CMCht and the

isothyocianate group of FITC (thiourea bond). Firstly, a 10 mg/mL CMCht/PAMAM dendrimer

nanoparticles solution was prepared in a carbonate/bicarbonate coupling buffer of pH 9.2. Then,

Chapter IV – Materials and Methods

82

50 µL of the FITC/DMSO solution was added per each mL of CMCht/PAMAM dendrimer

nanoparticle buffered solution under agitation, and kept in dark at 4 °C for 8 hours. The FITC-

labeled CMCht/PAMAM dendrimer nanoparticles solution was dialyzed against ultrapure water in

order to remove unlinked FITC for 24 hours and filtered (pore size < 220 nm) in sterile and dark

conditions. The final product was obtained as an orange powder after freeze-drying for 7 days.

This process was applied in all the Chapters that involve dendrimer NP: Chapter V, Chapter

Chapter VII and Chapter VIII.

IV-4.5. Synthesis of YIGSR – CMCht/PAMAM dendrimer nanoparticles

The attachment of YIGSR peptide (Genscript Corp, Piscataway, NJ, USA) to the NPs was

performed via carbodiimide chemistry to form an amide bond between the carboxyl groups on

the NPs and the amine groups at the peptide N-terminal end group (Figure VI-7). Briefly,

CMCht/PAMAM 1 % (w/v) was prepared in MES buffer (2-(N-morpholino)ethanesulfonic acid),

and mixed with the appropriate amount of NHS-Sulfo, EDC and a 10 mg/mL peptide solution

(dissolved in MES buffer). The solution was stirred for 24 hours and the NPs were then recovered

by dialysis against ultrapure water for 48 hours, followed by freeze-drying process for 7 days.

Figure IV-7 - Schematic representation of dendrimer nanoparticles successive modifications with

carboxymethilChitosan and YIGSR peptide from laminin via carbodiimide chemistry.

This reaction was performed on Chapter VIII, when aiming at targeting dendrimer

nanoparticles towards HCT-116 colorectal cancer cells, using YIGSR laminin peptide.

Chapter IV – Materials and Methods

83

IV-5. HYDROGELS PREPARATION

Current research has validated that tumor microenvironments play pivotal roles in tumor

development and metastasis over various physical, chemical, and biological factors, including

ECM composition, matrix remodeling, oxygen tension, pH, cytokines, and matrix stiffness [83].

An emerging trend in cancer research involves the creation of engineered 3D tumor models using

bioinspired hydrogels that accurately recapitulate the native tumor microenvironment. With recent

advances in materials engineering, many researchers are developing engineered tumor models,

which are promising platforms for the study of cancer biology and for screening of therapeutic

agents for better clinical outcomes [84].

Hydrogels are three-dimensional polymeric networks filled with water [85]. Its application for

several TE applications involve cell encapsulation strategies or drug delivery systems [86]. Typical

composition of hydrogels varies from synthetic (e.g. polyethylene glycol (PEG), polyacrylamide

(PAA), polydimethylsiloxane (PDMS)) to natural polymers (e.g. collagen, gelatin, alginate,

hyaluronic acid (HA), and chitosan). The gelation of hydrogels is enabled by either physical or

chemical cross-link methods. Physical gelation is possible through weak interactions between

polymer networks, whereas chemical cross link forms strong bonds between polymer chain [87].

The enzymatic crosslinking reactions can be advantageous for several reasons, including the

fact that this reaction process occurs at physiological conditions, without need any external

stimuli and taking only a few minutes to induce gelation. From the different enzymatic

crosslinking methods applied for TE purposes, it was reported that water-soluble polymers

containing phenol groups, aminophenol, tyramine or tyrosine can be crosslinked by the

horseradish peroxidase (HRP)/hydrogen peroxide (H2O2) complex system [88]. In this sense, the

SF protein can take advantage of its ∼5 % tyrosine groups to form enzymatically crosslinked SF

hydrogels via a peroxidase-mediated system [89, 90]. This is another goal of this thesis, i.e. to

take advantage of the easy processing of enzymatically crosslinked SF in order to create hydrogel-

based systems for TE and cancer research applications.

SF hydrogels previously proposed were prepared from a conformational transition of the SF

protein from random coil to β-sheet, using extreme physical treatments [91-93] or chemical

reagents [94], which in many cases involved long gelation times and harsh preparation

conditions that limited their use as ECM models. At the same time, these hydrogels showed

Chapter IV – Materials and Methods

84

homogeneous and highly tunable physicochemical properties, allowing for the modulation of

matrix stiffness [95].

Another purpose of this thesis was to take advantage from the peroxidase-mediated

crosslinked SF hydrogels and apply it to build a silk hydrogel microfluidic platform.

IV-5.1. Production of SF hydrogels

Along this thesis, SF was purified from the Bombyx mori cocoons in order to extract the glue-

like protein sericin and wax. For that, 5 g of cocoons were previously cut and cleaned into several

pieces and boiled for 1 hour in 2 L of sodium carbonate solution (0.02 M). Then, the purified SF

was washed for 1 hour in 1 L of boiling distilled water, followed by washing in distilled water

several times. Afterwards, SF was dried in a clean place. In order to obtain aqueous SF solution,

5 g of purified SF were dissolved in 25 mL of lithium bromide solution (9.3 M) at 70 ºC for 1

hour, yielding a solution around 20 % (w/v) [96]. The solution was dialyzed in distilled water for 2

days, using a benzoylated dialysis tubing (Molecular weight cut-off (MWCO): 2 kDa), changing the

distilled water at least 3 times per day. After this, it followed a concentration in a 20 wt.%

poly(ethylene glycol) solution (20,000 g/mol) for at least 6 hours [97]. The dialysis tubing was

carefully washed in distilled water and the solution collected to a 50 mL centrifuge tube (Thermo

Fisher Scientific, Waltham, MA, USA) (Figure IV-8). The final concentration of the SF solution was

determined by weight measurement. For that, around 500 μL of SF solution were weighted (wet

weight) and dried overnight at 70 ºC in an over. Then, the dry weight of the sample was

determined and the concentration obtained by dividing the dry weight by the wet weight.

Figure IV-8 - Concentrated aqueous SF solution in a 50 mL Falcon. Scale bar: 1.5 cm.

The SF solution was stored at 4 ºC until further use. HRP solution (0.84 mg/mL) and (H2O2)

solution (0.36 wt.%) were both prepared in PBS solution. The SF solutions (pH ∼7.0) were

Chapter IV – Materials and Methods

85

diluted into 16 wt.% by PBS and used for the hydrogels preparation. Micropipettes M100 and

M1000 (Gilson, Middleton, WI, USA) and corresponding tips were used for preparing the reaction

mixtures.

Figure IV-9 - Micropipettes Gilson M100 and M1000, specifically designed for hydrogel use. Scale bar: 1 cm.

In Chapter VI, each mL of SF solution was mixed with 100 μL of HRP and 65 μL of H2O2

solutions and left at 37 ºC for the crosslinking. For the 3D dynamic migration assays performed

in Chapter VI, 2 % and 3 % eSF hydrogel discs were prepared by adding 30 μL of the mixture

solutions in a 6 mm diameter silicon mold and placed at 37 °C for crosslinking during 20

minutes. This mold size was used for the crosslinked hydrogel to fit in the Vena4™ micro-wells.

For the SF hydrogels supplemented with hVCAM-1 (10 %) (Prepotech, Portugal), the solution was

mixed with hVCAM-1 (20 μg/mL) and then placed in silicon molds for crosslinking for 20

minutes. Afterward, hydrogels were taken out of the molds, placed in Vena4™ micro-well and

sealed (Figure IV-10).

Chapter IV – Materials and Methods

86

Figure IV-10 – Vena4™ empty and with silk hydrogels placed on the microwells. Scale bar: 6 mm.

IV-5.2. Production of Matrigel® coating and hydrogel

In the same Chapter VI, Matrigel was suggested and applied as ECM control material in the

form of coating (as suggested by manufacturers) and hydrogel at stock concentration (as a direct

comparison with silk), to study its influence in cancer cell migration. Matrigel coating in outward

cell migration (modified Boyden chamber) were reconstituted from Matrigel (BD Biosciences) and

diluted (250 µg/mL in 0.01 M Tris) (pH 8.0), NaCl (0.7 %) (Laborspirit, Portugal), according to

manufacturer’s instructions. Diluted Matrigel was carefully added to the top of the membrane

and incubated at 37 °C for 2 hours. Then, the remaining liquid (coating buffer) was carefully

removed from the permeable support membrane without disturbing the layer of Matrigel. For the

3D hydrogels, Matrigel were added to the 6 mm diameter silicon molds (30 µL) and placed at

37 °C for 30 minutes. For the hydrogels supplemented with hVCAM-1 (10 %), Matrigel was mixed

with hVCAM-1 and then placed in silicon molds for crosslinking for 30 minutes.

IV-6. MICROFABRICATION

Microfabrication describes ‘cleanroom’ fabrication processes for engineering solid flat

substrates [98]. It includes photolithography, direct-write laser (e-beam, X-ray) lithography, wet

etching, reactive ion etching, film deposition (plasma, chemical vapour, sputtering) and bonding

(eutectic, anodic, thermal, adhesion) [99]. It is mostly used for micro-electronic circuit

fabrication, but also for ‘lab-on-a-chip’.

Chapter IV – Materials and Methods

87

IV-6.1. Photolithography

In this thesis, photolithography principles were used to fabricate the microfluidic platforms in

Chapter VII and IX. Photolithography, also called optical lithography or UV lithography, is a

process used in microfabrication to pattern parts of a thin film or the bulk of a substrate (also

called a wafer)[100]. It uses light to transfer a geometric pattern from a photomask (also called

an optical mask) to a photosensitive chemical photoresist on the substrate. A series of chemical

treatments then either etches the exposure pattern into the material or enables deposition of a

new material in the desired pattern upon the material underneath the photoresist (Figure IV-11).

This method can create exceedingly small patterns, down to the nanometers size. It provides

precise control of the shape and size of the objects it creates and can create patterns over an

entire surface cost-effectively [101].

Figure IV-11 - Simplified illustration of photolithography microfabrication.

In Chapter VII the microfluidic device fabricated was designed using computer assisted

design software on a photolithography mask (Clewin 5). The designed microfluidic chip was

composed by three main compartments which consisted of: a large central chamber for an ECM

like hydrogel (5 mm diameter) and a pair of perfusable channels flanking the center chamber

(connected to the central chamber by a range of pillars to create superficial tension and avoid

hydrogels leakage into the lateral channels) (Figure IV-12). The microfluidic device in PDMS

(Down Corning) elastomer was fabricated by replica molding from a SU-8 (Microchem)/silicon

Chapter IV – Materials and Methods

88

master, based on a 10:1 ratio to curing agent. The master mold was produced by

photolithography. After pouring PDMS on mold, the mixture was degassed in a vacuum chamber

for air bubble removal, and cured at 80 °C for 1 hour in the oven. The devices were then cut out

by a razor blade, the fluidic connection ports punched, and bonding to a glass slide done after

oxygen plasma of both surfaces and conformal contact (Plasma cleaner/sterilizer Harrick®, high

RF level, 4 mbar, 120 seconds). The oxygen plasma treatment is a free-solvent method for

surface modification of polymer surfaces [102]. The plasma treatment has its only action at the

very top of structures, used to assemble the device to a glass slides. The PDMS microfluidic

network was irreversibly bonded to the glass substrate after both surfaces were treated with air

plasma.

Figure IV-12 - Clewin 5 software design of microfluidic chip with close up on microchannel/microwell interface.

IV-6.2. Soft-Lithography

Soft lithography is a field that covers a number of different techniques. Soft lithography can

be broken into three different major categories (each of which can be further divided up):

printing, replica molding and embossing. Soft lithography provides access to 3D and curved

structures, tolerates a wide variety of materials, generates well-defined and controllable surface

chemistries, and is generally compatible with biological applications. It is also low in cost,

experimentally convenient and has emerged as a technology useful for a number of applications

that include cell biology, microfluidics, lab-on-a-chip, micro-electromechanical systems and

flexible electronics/photonics [103]. In this thesis we have applied two of the sub-categories of

soft-lithography: Replica molding and microcontact printing (µCP).

Chapter IV – Materials and Methods

89

IV-6.2.1. Replica Molding

Replica molding is a process for shaping flexible polymer materials using a micro-sized rigid

frame or model called a 'mold'. A mold is a cast from the object to be replicated. This mold is

then filled with a pre-polymer, which is cross-linked and the resulting polymer is peeled off the

mold [104]. Using a mold, rather than the object itself, allows multiple copies to be made without

damaging the original.

This technique was applied in Chapter IX, where a silk hydrogel microfluidic device was

developed. The method for producing it also comprises photolithography as an initial step. As

proof of concept, a simple serpentine channel was designed using Autocad software. The silk

fibroin hydrogel-based microfluidic device was produced as follows: (i) the use of UV-

photolithography for the fabrication of microfluidic channels; (ii) replica moulding of the fabricated

structures using (PDMS; (iii) TCS-silanization and second replica moulding of the microfluidic

structures using a polymeric material (PDMS); (iv) replication of the microfluidic structures using

silk fibroin hydrogel (Figure IV-13).

Figure IV-13 – Example of serpentine like microchannels and stages of producing silk hydrogel microfluidic platform.

IV-6.2.2. Microcontact printing

Microcontact printing (μCP) is a soft lithography method whereby a PDMS stamp with a

desired motif is used to deposit molecules of interest onto a surface [105].

PDMS (1:10 w/w, Sylgard 184; Dow Corning) stamps were replicated from an SU-8 mold

(MicroChem) fabricated by UV-photolithography (MIDAS MDA-400MA). The PDMS stamp was

rendered hydrophilic by plasma treatment to maximize 67LR protein patterning. Glass coverslips

Chapter IV – Materials and Methods

90

were plasma treated for 1 minute and incubated for 1 hour under anoxic conditions in anhydrous

methanol solution of 5 % (v/v) in (3-aminopropyl)triethoxysilane (APTES). Next, the samples were

washed with methanol and ultrapure water. Then, APTES-functionalized coverslips were

incubated for 30 minutes in a solution of 1 % (v/v) glutaraldehyde in ultrapure water. The

micropatterned PDMS stamp was incubated with 10 μg/mL 67LR protein (Abcam, Portugal)

solution for 2 hours and the excess was removed by nitrogen drying. Immediately, the PDMS

stamp containing the laminin receptor protein was placed in contact with the functionalized glass

coverslip for 5 minutes. After releasing the stamp, FITC-labeled YIGSR-CMCht/PAMAM dendrimer

NPs were incubated overnight. This experiment was performed in order to prove the YIGSR

laminin protein attached to the FITC labeled NPs selectively binds to the patters containing 67LR

protein.

Figure IV-14 - µCP strategy applied in this thesis as proof-of-concept.

IV-6.2.3. Generation of stable gradients inside microfluidic chip

Chemical gradients play a key role in many biological processes and regulate a number of

cellular functions in vivo. Indeed, there are several examples of gradient-dependent phenomena

in nature, being a critical one the chemotaxis, i.e. the cell migration triggered by chemical

Chapter IV – Materials and Methods

91

gradients [106]. Moreover, chemical gradients have shown to have an impact various cell

behaviors, such as migration, proliferation, differentiation during development, wound healing,

inflammation, and tumorigenesis processes [107, 108].

Gradients generation can also have an impact in drug screening. The present state of

screening methods for drug discovery is still riddled with several ineptitudes [109]. Although

some widely used high-throughput screening platforms may enhance the drug screening process,

their cost and oversimplification of cell–drug interactions present a translational difficulty.

Microfluidic technology, and its advantage to use gradients may resolve many issues found in

conventional HTS technology. Moreover, they provide benefits such as reduced sample quantity

and integration of 3D cell culture physically more representative of the physiological/pathological

microenvironment [110].

In Chapter VII a gradient of nanoparticles was generated in the fabricated microfluidic

platform. For this, CMCht/PAMAM dendrimer nanoparticles were modified with a green

fluorescent molecule FITC as described before. Matrigel was injected into the central chamber

with a syringe and allowed to crosslink for 30 minutes at 37 ºC. After this, a source-sink

mechanism was created, with one solution of 0.5 mg/mL FITC-labeled nanoparticles in PBS

perfused through one side channel and PBS was perfused through the other inlet, using a Nexus

3000 syringe pump (Chemix). Syringes of 3 mL syringes were used to introduce PBS with or

without the NP at a rate of 8 μL/hour. The experiment was performed in a Nikon Eclipse Ti-E

fluorescence microscope (Nikon Instruments, Japan) for time lapse imaging of the gradient

formation. Profile and stability of the dynamic gradient was analyzed using NIS-Elements AR

(Nikon) software. The chips were analyzed at different time points between 0 and 12 hours with

continuous perfusion. Built-in “plot profile” module was used for assessing the gradient intensity

and stability over time. Figure IV–14 represents the process of placing the microfluidic platform

on the fluorescence microscope and analyze in real time the formation of a gradient of

fluorescent labeled dendrimer NPs.

Chapter IV – Materials and Methods

92

Figure IV-15 - A) Apparatus for generating gradients in real time. B) Microfluidic device under perfusion to create a

gradient. C) Fluorescence image of the FITC-CMCht/PAMAM nanoparticles’ gradient.

IV-7. PHYSICOCHEMICAL CHARACTERIZATION TECHNIQUES

IV-7.1. Morphological/morphometric evaluation

IV-7.1.1. Scanning electron microscopy (SEM)

Scanning Electron Microscopy (SEM), is an imaging technique that produces images of a

sample by scanning the surface with focused beam of electrons. When in contact with the atoms

at the sample’s surface, the electrons produce several signals containing the collected

information. Thus, the qualitative information regarding the surface morphology of samples and

elemental information can be obtained [111].

In Chapter VIII, the microstructure of the surface of L929 and HCT-116 cells were analyzed

on top of polystyrene coverslips in order to detect if the fluorescent signal of nanoparticles

obtained in confocal microscope was due to the presence of nanoparticles at the surface or

internalized at the cell’s cytoplasm, using scanning electron microscope, SEM (Leica Cambridge

S-360, UK). Prior to, specimens were coated with carbon (Fisons Instruments, Polaron SC 508,

UK) and gold (Fisons Instruments, Polaron SC 502, UK), respectively. The current was set at 18

mA with a coating time of 120 seconds.

In Chapter IX SEM was also applied to study the fidelity of the silk hydrogel microfluidic

platform. Samples were produced by enzymatic crosslinking as described before, and then dried

in a sequence of ethanol solution from 10 % to 100 % (v/v). After dehydration, samples were

Chapter IV – Materials and Methods

93

dried using critical point dryer (Autosamdri-815 Series A, Tousimis, USA). Using such approach,

the samples were heated up to 37 ºC and pressurized until 90 bar for 45 minutes.

Figure IV-16 - SEM with EDS analyzer. The SEM (JSM-6010 LV, JEOL, Japan) instrument is equipped with the low

vacuum mode as a standard feature. The low vacuum mode allows non-conductive specimens to be

observed and analyzed without conductive coating. This SEM is equipped with an energy dispersive

spectroscope (EDS). Scale bar: 10 cm.

IV-7.1.2. Fourier Transform Infrared Spectroscopy (FTIR)

Fourier Transform Infrared Spectroscopy (FTIR) is a cost-effective technique. It is a form of

vibrational spectroscopy that is useful in the study of a variety of chemical processes. It identifies

polymers and analyzes polymer chemical modifications. Fourier transformation algorithm allied to

IR spectroscopy gives a spectrum of IR absorption per frequency/wavelength. Similar chemical

groups absorb in the IR at similar frequencies, enabling to identify the chemical structure of a

compounds, and to identify chemical modifications [112]. IR spectrum is specific for each

compound. FTIR was performed to verify the presence of covalently liked carboxymethylchitosan

in dendrimer NP in Chapter V, as well as the chemical modification with YIGSR peptide in

Chapter XIII. A transmittance spectrum was obtained on an IR Prestige-21 FTIR spectrometer

(Shimadzu) by performing 32 scans in each spectrum over a range of 400-4400 cm-1 and with a

resolution of 4 cm-1. In Chapter IX, the silk hydrogel protein conformation was evaluated by an

Attenuated Total Reflectance (ATR) model (IRPrestige-21, Shimadzu, Japan) in a FTIR equipment

Chapter IV – Materials and Methods

94

(Perkin-Elmer 1600 series equipment, CA, USA) equipped with a Germanium crystal, as can be

seen in Figure IV-16. All spectra were obtained between 400-4400 cm−1, at a 4 cm−1 resolution for

an average of 50 scans.

Figure IV-17 - The Fourier transform infrared (FTIR) spectrometer is used as a standard chemical characterization

technique. The samples can be analyzed as films, KBr discs, powder or liquids. The available

methodologies comprise transmittance, specular reflectance, diffuse reflectance and attenuated total

reflectance (ATR). Scale bar: 10 cm.

IV-7.1.3. Transmission Electron Microscopy (TEM)

Transmission Electron Microscopy (TEM) is a damaging imaging system that provides high-

resolution images. The principle of the technique consists of electron beams that pass through

the samples and interact with the atoms at different depths. The transmitted electrons originate

signals which are detected by the TEM equipment and converted into a two-dimensional image.

Importantly, TEM’s resolution is about 0.2 nm, which is more than a thousand-fold greater than

the resolution of the light microscope [113]. In Chapter V, a JEOL JEM 1400 TEM (Tokyo, Japan)

equipment was used to investigate and measure the CMCht/PAMAM dendrimer NP.

In Chapter VIII, TEM was used to compare and analyze CMCht/PAMAM dendrimer

nanoparticles with and without the peptide YIGSR attachment, since an increase in size would be

expected. For the analysis, samples were mounted on a 200-mesh copper or nickel grids, stained

with 2 % Phosphotungstic acid and examined. Images were digitally recorded using a CCD digital

camera (Orious 1100W Tokyo, Japan).

Chapter IV – Materials and Methods

95

IV-7.1.4. Atomic Force Microscopy (AFM)

Atomic Force Microscopy (AFM) technique allows the observation and measurement of the

surface structure of samples with an unprecedented resolution and accuracy. Through this

analysis, it is possible to obtain images showing the arrangement of individual atoms in a surface

sample, or to see the structure of individual molecules [114]. Comparing to the optical or

electron microscopes, AFM does not form an image by focusing light or electrons onto a surface.

It “senses” the samples surface with a sharp probe or cantilever and builds a map covering the

height or topography of the samples surface. The type of images generated by this microscope

can reach a magnification greater than 1000 x. On the other hand, from an imaging microscope,

only a 2D projection of a samples surface can be reached without any height information

included. Furthermore, besides being an imaging tool AFM has several spectroscopic modes that

measure other samples properties as nanoscale [114]. In Chapter V the morphology of the NPs

was investigated using AFM available in the research facilities (Figure VI–17). First, the freeze-

dried CMCht/PAMAM dendrimer NPs were dispersed in ultrapure water to obtain a solution with

final concentration of 1 mg/mL and then one drop was placed over a 9.9 mm mica disc (Agar

Scientific, Essex, England) and blown dried with nitrogen gas for subsequent characterization.

Then, the samples were analyzed using the Tapping Mode™ with a MultiMode AFM connected to

a NanoScope III controller, both from Veeco (NY USA), with noncontact silicon nanoprobes (ca.

300 kHz) from Nanosensors, Switzerland. All images were plane-fitted using the third-degree-

flatten procedure included in the Nano-Scope software version 4.43r8. The morphometric

analysis was performed using the same software.

In Chapter VIII, AFM analysis was performed using the same exact procedure described for

Chapter V, this time in order to compare CMCht/PAMAM dendrimer NPs to YIGSR modified

CMCht/PAMAM dendrimer NPs.

Chapter IV – Materials and Methods

96

Figure IV-18 - The AFM is able to acquire images of flat surfaces that can encode the surface topography,

mechanical response, among other properties. The equipment presents a series of accessories that

allows the acquisition of data in liquids and air. In the case of liquids, it is also possible to control the

temperature of the experiments. Scale bar: 10 cm.

IV-7.1.5. Differential scanning calorimeter (DSC)

Differential scanning calorimeter (DSC) is an important instrument in thermal analysis, being

useful to understand amorphous and crystalline behavior, eutectic transitions, curing and degree

of cure, and many other material’s properties used to design, manufacture, and test products.

This technique allows to study the material’s heat capacity (Cp) as function of the temperature. A

sample of known mass is heated or cooled and the changes in its Cp are tracked as changes in

the heat flow, which allows the detection of transitions such as melting, glass transitions, among

others parameters [115]. The DSC experiments performed in this thesis were carried out using

DSC Q100 V9.8 Build 296 apparatus, which can be seen in Figure IV-18. In Chapter V, we

decided to study the thermal stability and changes in crystallinity over a range of temperatures of

the developed CMCht/PAMAM dendrimer nanoparticles. For that, the samples were placed in

aluminum pans and heated at a rate of 10 °C/min from 20 to 220 °C, cooled to 20 °C and

heated at 5 °C/minute until 200 °C. Standard calibrations were performed using indium leads.

Chapter IV – Materials and Methods

97

Figure IV-19 - Differential scanning calorimetry is a thermal analysis technique that detects the temperatures and

heat flows caused by changes in heat capacity or by endothermic and exothermic processes of

materials as a function of time and temperature. Scale bar: 10 cm.

IV-7.1.6. Nuclear Magnetic Resonance Spectroscopy (NMR)

Nuclear magnetic resonance (NMR) is a phenomenon which occurs when the nuclei of

certain atoms are immersed in a static magnetic field and exposed to a second oscillating

magnetic field. Some nuclei experience this phenomenon, and others do not, dependent upon

whether they possess a property called spin. H1NMR is the application of nuclear magnetic

resonance in NMR spectroscopy with respect to hydrogen nuclei within the molecules of a

substance. This is a non-destructive technique that allows determining the structure of organic

compounds. Simple NMR spectra are recorded in solution, and solvent protons must not be

allowed to interfere. In this thesis, H1NMR analysis was performed twice, first to investigate the

structure and the successful modification of PAMAM dendrimers with CMCht in Chapter V, and

then to assess the covalent link of YIGSR peptide to CMCht/PAMAM in Chapter VIII. For this

purpose, nanoparticles were dissolved in deuterated water (D2O). The NMR spectra were obtained

with a Mercury -400BB operating at a frequency of 399.9 MHz at 50 °C. The one-dimensional

1H spectra were acquired using a 45° pulse, a spectral width of 6.3 kHz and an acquisition time

of 2.001 seconds.

Chapter IV – Materials and Methods

98

IV-7.1.7. Zeta potential and particle size analysis

Adequate characterization of NPs is of paramount importance to develop well-defined

nanoformulations of therapeutic relevance. Determination of particle size and surface charge of

NPs are indispensable for proper characterization of NPs. DLS (dynamic light scattering) and ZP

(zeta potential) measurements have gained popularity as simple, easy and reproducible tools to

ascertain particle size and surface charge [116].

Almost all particulate or macroscopic materials in contact with a liquid acquire an electronic

charge on their surfaces [117]. ZP is an important and useful indicator of this charge which can

be used to predict and control the stability of colloidal suspensions or emulsions, for example.

The greater the zeta potential the less likely the suspension is to be aggregated. On the other

hand, particle size can be determined by measuring the random changes in the intensity of light

scattered from a suspension or solution. This technique is commonly known as DLS, but is also

called photon correlation spectroscopy (PCS) and quasi-elastic light scattering (QELS). ZP and

particle size of the CMCht/PAMAM dendrimer nanoparticles and YIGSR-CMCht/PAMAM

dendrimer nanoparticles were measured in a particle size analyzer (Zetasizer Nano ZS, Malvern

Instruments, UK) (Figure IV–19) in Chapters V and VIII. By analyzing the size of particles and

surface charge, important considerations can be made on the success of engineering these

particles.

Chapter IV – Materials and Methods

99

Figure IV-20 - Zetasizer Nano ZS, Malvern Instruments. Scale bar: 10cm.

ZP was calculated by determining the electrophoretic mobility and then applying the Henry

equation. The electrophoretic mobility is obtained by performing an electrophoresis experiment

on the sample and measuring the velocity of the particles using Laser Doppler Velocimetry (LDV).

The Henry equation is:

Equation IV-3 – Henry equation.

z: Zeta potential;

UE: Electrophoretic mobility;

ε: Dielectric constant;

η: Viscosity;

ƒ(Ka): Henry’s function; the value used was 1.5 and is referred to as the Smoluchowski

approximation.

Chapter IV – Materials and Methods

100

Electrophoretic determinations of ZP were investigated using the universal ‘dip’ cell, using H2O

and phosphate buffered saline (PBS) solutions pH 7.4.

IV-7.1.8. Micro-BCA

Micro BCA™ Protein Assay Kit is a detergent-compatible bicinchoninic acid formulation for

the colorimetric detection and quantitation of total protein. A purple-colored reaction product is

formed by the chelation of two molecules of BCA with one cuprous ion (Cu+1). This water-soluble

complex exhibits a strong absorbance at 562 nm that is linear with increasing protein

concentrations. Moreover, the macromolecular structure of protein, the number of peptide bonds

and the presence of four amino acids (cysteine, cystine, tryptophan and tyrosine) are reported to

be responsible for color formation [118]. In Chapter VIII, Micro BCA Kit (Thermo Fisher) was

used to measure the protein concentration of grafted YIGSR peptide onto CMCht/PAMAM

dendrimer nanoparticles. Since the peptide sequence YIGSR incorporates the amino acid

tyrosine, this technique was chosen for protein concentration quantification. Samples of

CMCht/PAMAM and modified YIGSR-CMCht/PAMAM dendrimer NPs were dissolved in ultrapure

water at a final concentration of 1 % and compared to standards for protein quantification in

triplicates using the microplate reader (Synergy HT, BioTek Instruments, USA) at 562 nm.

IV-7.1.9. Mechanical properties - Rheological analysis

Rheology is the study of flow and deformation of materials under applied forces which is

routinely measured using a rheometer. The measurement of rheological properties is applicable

to all materials from fluids such as dilute solutions of polymers and surfactants through to

concentrated protein formulations, to semi-solids such as pastes and creams, to molten or solid

polymers as well as asphalt [119]. This technique gives the elasticity, viscosity and plastic

behavior of materials under changes of strain, frequency, time, or temperature. Moreover, the

existence of any chemical reactions in the flows that induce gelation or polymerization can also

be detected with high sensitivity through this technology.

In Chapter VIII, a rheological experiment were performed to characterize the SF and Matrigel

hydrogels using a Kinexus pro+ rheometer (Malvern Instruments, UK), equipped with the

Chapter IV – Materials and Methods

101

acquisition software rSpace (Figure IV-20). SF hydrogel discs were first prepared in PDMS

silicone molds (8 mm diameter and 2 mm height), and after each time point were analyzed by

oscillatory experiments. For that, the measuring system was equipped with stainless steel (316

grade) parallel plates: an upper measurement geometry plate (8 mm diameter) and a lower

pedestal (20 mm diameter) with roughened finish. The influence of polymer type and

concentration of the used hydrogels was studied in terms of their rheological behavior in order to

analyze the mechanical properties of the hydrogels. Stress sweeps (0.1 Hz) were first performed

to determine the linear viscoelastic region (LVR) for all the tested conditions. Frequency sweeps

(0.01–10 Hz) were then performed within the LVR. The values of the shear storage modulus (G′)

and loss modulus (G″) were obtained at a frequency of 0.1 Hz. Storage modulus G′ is known as

the “solid-like” or elastic component of the gel, and G″ (Loss modulus) as the “liquid like” or

viscous component. The system is considered a gel if the value of G′ is greater than G″, which

was expected since the samples being analyzed were hydrogels.

Chapter IV – Materials and Methods

102

Figure IV-21 – Rheometer apparatus, Malvern Instruments. Scale bar: 10 cm.

IV-8. IN VITRO BIOLOGICAL TESTING

IV-8.1. Cell sources

Most of the cellular work performed in this thesis was carried out using cell lines. Cells lines

are immortalized cells that present the ability to proliferate indefinitely either due to a random

mutation or due to a programmed modification. Cell lines are useful models for doing research,

because they provide reliability in experimental results due to the possibility to obtain large

amounts of cells for prolonged use [120].

Furthermore, since this thesis was developed in the context of cancer research,

immortalized cancer cells were abundantly used.

IV-8.1.1. HCT-116 cells

HCT-116 cancer cell line was purchased from ATCC (USA), (Ref. ATCC® CCL-247). Cells

were cultured in low glucose Dulbecco’s modified Eagle’s medium (DMEM), supplemented with

44 mM of sodium bicarbonate, 10 % (v/v) of fetal bovine serum (FBS; Biochrom, Berlin,

Chapter IV – Materials and Methods

103

Germany), and 1 % (v/v) of antibiotic/antimycotic solution (prepared with 10,000 units/mL

penicillin G sodium, 10.000 μg/mL streptomycin sulfate, and 25 μg/mL amphotericin B as

Fungizone (R) in 0.85 % saline; Gibco®, Life Technologies, Carlsbad, CA, USA). Cells were

routinely trypsinized (0.25 % trypsin–EDTA solution; Life Technologies, Carlsbad, CA, USA) for 2-

3 minutes at 37 ºC, centrifuged at 300 g for 5 minutes (Eppendorf 5810R centrifuge equipped

with an A-4-62 rotor), and re-suspended at a density of 1x106 cells in T150 cell culture flasks (BD

Biosciences, Enzifarma, Porto, Portugal).

IV-8.1.2. HeLa cells

HeLa cell line (from cervical cancer cells) was generously donated by Prof. Joseph Costello

(California University, Neurosurgery Department, San Francisco, USA) and further provided by

Prof. Rui M. Reis (Life and Health Science Research Institute, University of Minho, ICVS/3B’s –

PT Government Associate Laboratory, Portugal). Cells were cultured in low glucose Dulbecco’s

modified Eagle’s medium (DMEM), supplemented with 44 mM of sodium bicarbonate, 10 % (v/v)

of fetal bovine serum (FBS; Biochrom, Berlin, Germany), and 1% (v/v) of antibiotic/antimycotic

solution (prepared with 10,000 units/mL penicillin G sodium, 10.000 μg/mL streptomycin

sulfate, and 25 μg/mL amphotericin B as Fungizone (R) in 0.85 % saline; Gibco®, Life

Technologies, Carlsbad, CA, USA). The L929 cells were incubated in an atmosphere containing 5

% CO2 at 37 ºC, and the medium changed every 2-3 days. Cells were routinely trypsinized (0.25

% trypsin–EDTA solution; Life Technologies, Carlsbad, CA, USA) for 2-3 minutes at 37 ºC,

centrifuged at 300 g for 5 minutes (Eppendorf 5810R centrifuge equipped with an A-4-62 rotor),

and re-suspended at a density of 1X106 cells in T150 cell culture flasks (BD Biosciences,

Enzifarma, Porto, Portugal).

IV-8.1.3. GBM cells

Human neuronal glioblastoma U87MG cell line was generously donated by Prof. Joseph

Costello (California University, Neurosurgery Department, San Francisco, USA) and further

provided by Prof. Rui M. Reis (Life and Health Science Research Institute, University of Minho,

ICVS/3B’s – PT Government Associate Laboratory, Portugal). The use of U87MG cell line was

Chapter IV – Materials and Methods

104

approved by the Ethics Committee of University of Minho. Cells were cultured in low glucose

Dulbecco’s modified Eagle’s medium (DMEM), supplemented with 44 mM of sodium

bicarbonate, 10 % (v/v) of fetal bovine serum (FBS; Life Technologies, Carlsbad, CA, USA), and 1

% (v/v) of antibiotic/antimycotic solution. The U87MG cells were incubated in an atmosphere

containing 5 % CO2 at 37 ºC, and the medium changed every 2-3 days. Cells were routinely

trypsinized (TryplE™ Express) for 5 minutes at 37 ºC, centrifuged at 300 g for 5 minutes, and re-

suspended at a density of 1X106 cells in T75 cell culture flasks.

IV-8.1.4. L929 cells

L929 cell line from mouse C3H/An connective tissue (purchased from ECACC, UK) was

cultured in low glucose Dulbecco’s modified Eagle’s medium (DMEM), supplemented with 44

mM of sodium bicarbonate, 10 % (v/v) of fetal bovine serum (FBS; Biochrom, Berlin, Germany),

and 1 % (v/v) of antibiotic/antimycotic solution (prepared with 10.000 units/mL penicillin G

sodium, 10.000 μg/mL streptomycin sulfate, and 25 μg/mL amphotericin B as Fungizone (R) in

0.85% saline; Gibco®, Life Technologies, Carlsbad, CA, USA). The L929 cells were incubated in

an atmosphere containing 5 % CO2 at 37 ºC, and the medium changed every 2-3 days. Cells were

routinely trypsinized (0.25% trypsin–EDTA solution; Life Technologies, Carlsbad, CA, USA) for 2-3

minutes at 37ºC, centrifuged at 300 g for 5 minutes (Eppendorf 5810R centrifuge equipped with

an A-4-62 rotor), and re-suspended at a density of 1X106 cells in T150 cell culture flasks (BD

Biosciences, Enzifarma, Porto, Portugal).

IV-8.1.5. HCOMECS cells

Human Colonic Microvascular Endothelial Cells were isolated by ScienCell research

laboratories from human colonic tissue and purchased from Innoprot (Ref. P10769, The

Netherlands) cryopreserved (liquid nitrogen) at a cell density of 5x105 cells/vial (1 mL) as primary

cultures and delivered by the company frozen. The cell content in each vial was dispensed into a

T75 cell culture flask at a density of 5x103 cells/cm2 and cultured in endothelial medium (Ref.

P60104, purchased from Innoprot, The Netherlands). Culture medium was changed every 3

days, until reaching approximately 70 % confluence. Cells were routinely trypsinized (0.25 %

Chapter IV – Materials and Methods

105

trypsin–EDTA solution) for 2-3 minutes at 37 ºC, centrifuged (200 g, 5 minutes) and re-

suspended at a density of 1x106 cells in T150 cell culture flasks.

IV-8.2. Cryopreservation

The cryopreservation of the cells used in this thesis was performed using a Statebourne

Biosystem 24 cryogenic tank (Statebourne Cryogenics Ltd., UK). Briefly, cell suspensions of

1x106 cells/mL were prepared in a cryopreservation solution, consisting of 10 % (v/v) DMSO

(N182, VWR, Radnor, PA, USA) in FBS, and transferred into 1.5 mL cryovials (479-6841, VWR,

Radnor, PA, USA). Then, cell suspensions were gently cooled down, first at -20 ºC for at least 2

hours and then moved to -80 ºC freezer for a minimum period of 12 hours. The cryovials were

subsequently stored at -176 ºC in the gas nitrogen phase of the cryogenic tank.

IV-8.3. Cell seeding techniques

IV-8.3.1. Seeding on Vena8™ chips

In Chapter V a variety of cancer cells lines including HeLa cells, HCT-116 cells and GBM

cells were seeded in Vena8™ microfluidic chip in order to compare the effect of CMCht/PAMAM

nanoparticles in terms of cytotoxicity in static (2D culture flasks) and dynamic (2D under flow).

For this, a semi-automated microfluidic platform, Vena8™ biochip with a glass coverslip (Tebu-

Bio, Portugal), was used to mimic physiological flow conditions though the perfusion of cell

culture medium in the microchannels. Each chip microchannel was coated with fibronectin

(Sigma, Germany) and then placed in a humidified sterile Petri dish, incubated at 37 °C for 1.5

hours. After the incubation period, the confluent cells were detached from the T150 cell culture

flasks using trypsin (0.25 % trypsin–EDTA solution (v/v)), centrifuged at 300 g for 5 minutes,

counted in a cell counter KOVA® Glasstic® slide 10 chamber with grids (Garden Grove, CA,

USA) and diluted in a new cell suspension. For the seeding, 15.000 cells (HeLa, U87MG or HCT-

116) were gently added into each microchannel and kept in a sterile Petri dish in the incubator

for 2 hours to allow cell adhesion. To prevent drying, after the initial 30 minutes of incubation,

complete medium DMEM was added to the channels.

Chapter IV – Materials and Methods

106

IV-8.3.2. Seeding on Vena4™ chips

In Chapter VI this thesis starts to focus only on colorectal cancer models. Therefore, HCT-

116 cells were seeded on Vena4™ chips’ microchannel in order to evaluate the influence of SF

and Matrigel’s mechanical properties, as well as the influence of hVCAM-1 molecule in cell

migration. All the procedures were performed under aseptic condition. Confluent cells were

detached from the T150 cell culture flasks using trypsin (0.25 % (v/v) trypsin–EDTA solution),

centrifuged at 300 g for 5 minutes, and diluted in a new cell suspension in PBS for labeling with

CellTracker™ CM-DiI fluorescent dye. This dye has been designed to freely pass through cell

membranes into cells, where it is transformed into cell membrane-impermeant reaction products.

CellTracker™ CM-DiI dye is retained in living cells through several generations. It is stable,

nontoxic at working concentrations, well retained in cells, and brightly fluorescent at physiological

pH. HCT-116 cells were pre-labeled with red cell tracker CM-DiI Dye (1x10−6 M in PBS)

(Invitrogen, Portugal) and incubated at 37 °C for 30 minutes. After this, labeled cells were

centrifuged again at 300 g for 5 minutes, counted in a cell counter KOVA® Glasstic® slide 10

chamber with grids (Garden Grove, CA, USA) and seeded on Vena4™ microchannels using a

standard yellow pipette (30.000 cells/channel). In order to prevent drying, after the initial 30

minutes of incubation, complete medium DMEM was added to the channels.

IV-8.3.3. Seeding on fabricated microfluidic chip

Our goal in Chapter VIII was to develop a complex 3D microfluidic chip-based in vitro model

that could emulate the human colorectal tumor microenvironment. In this sense, the designed

and fabricated microfluidic device has three compartments: a circular central chamber for the

ECM like hydrogel, (where HCT-116 cancer cells are encapsulated in Matrigel to mimic the

tumor), laterally sided by a pair of perfused channels. The two lateral channels are not

interconnected between them, and are used to mimic the vascular part of the tumor by seeding

colonic endothelial cells.

Chapter IV – Materials and Methods

107

IV-8.3.3.1 Encapsulation of HCT-116 in Matrigel

In Chapter VIII HCT-116 cancer confluent cells were detached from the T150 cell culture

flasks using trypsin (0.25 % (v/v) trypsin–EDTA solution), centrifuged at 300 g for 5 minutes,

diluted in a new cell suspension and counted in a cell counter KOVA® Glasstic® slide 10

chamber with grids (Garden Grove, CA, USA). After this, HCT-116 cells were re-suspended in

Matrigel supplemented with VEFG (100 ng/mL) (R&D Systems) in a density of 10x106 cells/mL,

and injected in the central chamber of the microfluidic device using a syringe. It was let to

incubate for 15 minutes at 37 ºC for crosslinking.

IV-8.3.3.2 Seeding of HCOMECS in lateral microchannels

The lateral channels were pre-coated using a 10 % (v/v) Matrigel solution in DMEM complete

medium and incubated for 30 minutes at 37 ºC. HCoMECs confluent cells were detached from

the T150 cell culture flasks using trypsin (0.25 % (v/v) trypsin–EDTA solution), centrifuged at 300

g for 5 minutes, diluted in a new cell suspension and counted in a cell counter KOVA® Glasstic®

slide 10 chamber with grids (Garden Grove, CA, USA). Using the same the same density as for

the cancer cells (10x106 cells/mL), cells were seeded on the lateral channels using a yellow

pipette and left to adhere for 6 hours. Then, cells were exposed to a perfusion flow regime at a

flow rate of 8 μL/hour for the duration of the experiments. At days 1 and 5 days, the cell culture

was evaluated using bright field/fluorescence microscopy.

IV-8.4. Cell migration assays

Migration is often used as broad term in biology to describe any directed cell movement

within the body. The ability to migrate allows cells to change their position within organs or

tissues [121]. In pathology, invasion of carcinomas for example is defined as the penetration of

tissue barriers, such as passing the basement membrane and infiltration into the underlying

interstitial tissues by malignant tumor cells [122]. Intestinal cancers for example are classified as

invasive when the tumor mass has crossed the basal membrane and entered the submucosal

muscle layer. Therefore, migration can occur on 2D surfaces without any obstructive fiber

network. However, invasion is defined as cell movement through a 3D matrix, which is generally

Chapter IV – Materials and Methods

108

complemented by a restructuring of the surrounding 3D environment [44]. Determining the

migratory and invasive capacity of tumor cells and clarifying the underlying mechanisms is of

upmost relevancy for novel strategies in cancer diagnosis, prognosis, drug development and

treatment. As such, there are several migration/invasion assays described in the literature [123],

some of which were performed in this thesis, as follows.

IV-8.4.1. Outward Cell Migration (Modified Boyden Chamber Assay)

The transwell assay was originally introduced by Boyden (often called Boyden chamber

assay) to analyze the chemotactic responses of leukocytes [123]. The principle of this assay is

based on two medium containing chambers separated by a porous membrane through which

cells transmigrate. Generally, cells are seeded in the upper part and can migrate in vertical

direction through the pores of the membrane and into the lower compartment. This lower

compartment usually contains an attractant or simply higher serum [124]. Alternatively, and

similar to what was used in Chapter VI, dark colored porous membranes are available, which

block light transmission (FluoroBlok, Becton Dickinson) from unmigrated cells. Therefore, the

detection and quantification of the migrated labeled cells is simplified and there is no need to

remove the remaining cells from the top-side of the membrane. In Chapter VI this assay was

modified and adapted to our needs (Image IV–21), in order to compare the migration results to a

microfluidic dynamic platform. For this, the upper and lower compartments were separated not

only by a porous black membrane, but also by a layer of the proposed studied hydrogels. First,

the SF hydrogels and Matrigel coating were produced as described before. Then, HCT-116 cells

were pre-labeled with Cell Tracker Red (5x10−3 M) (Invitrogen, Portugal) for 20 minutes, and then

seeded on top of the coating/hydrogels (30 000 cells/well). In the upper chamber, either DMEM

medium with 10 % (v/v) FBS or Serum free medium were added (300 μL). In the lower chamber,

either DMEM medium with 10 % (v/v) FBS (control), DMEM supplemented with 10 % (v/v) h-

VCAM1 or with 30 % (v/v) FBS were added (500 μL). At different time points, fluorescence

intensity from the bottom was measured using a microplate spectrofluorimeter (BioTek, Portugal)

in area-scan bottom-reading mode, at excitation/emission wavelengths of 553/570 nm. Results

are presented as the increase in fluorescence in relation to the time point 0 hours. For qualitative

analysis, in the last time point (48 hours), cells were fixed with formalin (10 % v/v) (Sigma,

Chapter IV – Materials and Methods

109

Germany) and stained for F-actin filaments of the cytoskeleton and nuclei with Texas Red-X

phalloidin (Molecular Probes, Invitrogen, USA) and with 4,6-diamidino-2-phenylindole, dilactate

(DAPI blue, Molecular Probes, USA), respectively, following supplier’s protocol. Fluorescence

images of migrating cells were collected using an inverted fluorescence microscope (Leica,

Germany).

Figure IV-22 – Schematics of Modified Boyden Chamber assay with the experimental conditions used in Chapter VI.

IV-8.4.2. Cell Migration Assay in 3D VenaT4 Biochips

In Chapter VI, cancer cell migration was tested and compared in the Modified Boyden

Chamber invasion assay and in the 3D platform Vena4™ (Cellix, Ireland).

Each chip’s microchannel was coated with fibronectin (Sigma, Germany) (100 μg/mL),

placed in a humidified sterile Petri dish and incubated at 37 °C for 1.5 hours. After the

incubation period, HCT-116 confluent cells were detached from the T150 cell culture flasks using

trypsin (0.25 % (v/v) trypsin–EDTA solution), centrifuged at 300 g for 5 minutes, diluted in a new

cell suspension and counted in a cell counter KOVA® Glasstic® slide 10 chamber with grids

(Garden Grove, CA, USA). Cells were pre-labeled with red cell tracker CM-DiI Dye (1x10−6 M)

(Invitrogen, Portugal) and incubated at 37 °C for 30 minutes. After, labeled cells were seeded on

Vena4™ microchannels using a standard yellow pipette (30,000 cells/channel). In order to

prevent drying, after the initial 30 minutes of incubation, complete medium DMEM was added to

the channels. The hydrogels (3 % and 2 % SF and Matrigel), previously crosslinked in the silicon

molds, were placed in the microwell and sealed. At determined time points (0 and 48 hours), the

Chapter IV – Materials and Methods

110

chip was observed under confocal laser scanning microscopy (Leica, Germany) to monitor pre-

labeled cells’ migration toward the hydrogel with and without hVCAM-1.

IV-8.4.3. Endothelial cell Invasion

Cancer has the ability to spread to adjacent or distant organs, which makes it life

threatening. Tumor cells can penetrate blood or lymphatic vessels, circulate through the

intravascular stream, and then proliferate at another site: metastasis. For the metastatic spread

of cancer tissue, angiogenesis is important [125]. Angiogenesis is a complex morphogenetic

process where endothelial cells from pre-existing vessels invade the matrix as sprouts to form

new vessels [125]. For capillary sprouting and blood vessel formation, both proliferation and

migration are essential, but how these two events are coordinated is not fully clear [126].

Angiogenesis occurs in several well-characterized stages: At first, stimulatory signals bind to

surface receptors on vascular endothelial cells. Activated endothelial cells then release proteases

that degrade the underlying basement membrane and allow endothelial cells to migrate out of

the existing blood vessel. The endothelial cells begin to proliferate and form sprouts that extend

toward the source of the angiogenic stimulus. Subsequent lumenation of these primary sprouts

leads to formation of capillary loops, which is followed by synthesis of a new basement

membrane and blood vessel maturation to complete tube-like structures through which blood can

flow. Pro-angiogenic signals, such as VEGF, are known to promote processes such as

proliferation, endothelial motility, and filopodia extension [127].

Angiogenesis inhibitors can be designed to block the formation of new blood vessels, and

the growth of tumors would thereby be halted but not eliminated; hence, anti-angiogenesis

monotherapies are not effective in humans as was hoped for. Thus, combinatorial treatments

with conventional chemotherapy drugs are required. These inhibitors sometimes may not

eliminate tumors and in order to achieve optimal treatment, a combination of anti-angiogenesis

and conventional chemotherapy may be required [125]

In Chapter VII, the 3D formation of endothelial sprouts in a microfluidic device and the

changes in cellular organization during early stages of invasion were examined. Seeding of

colonic microvascular endothelial cells on fabricated microfluidic chip was performed as

described in section “Seeding of HCOMECS in lateral microchannels”. During cell culture, bright

Chapter IV – Materials and Methods

111

field images were taken with inverted microscope to analyze and quantify in which extent

endothelial cells remain on the lateral microchannels or respond to VEGF present in the Matrigel

in the central chamber (Figure IV-22). Three images of each time point (day 0, day 2, day 3 and

day 5) were analyzed. The length was measured using Image J (Fiji), from the outer limit of the

microchannels/hydrogel interface, until the furthest point of the elongated cell(s).

Figure IV-23 - Scheme representation of endothelial cell invasion from lateral channels towards the central chamber

following VEGF gradient. Scale bar: 1 mm.

IV-8.5. Metabolic activity and cell viability examination

IV-8.5.1. Alamar blue assay

Alamar blue assay is used to evaluate the metabolic activity of cells. It is a cell viability

reagent that works by using the reducing power of living cells to quantitatively measure their

metabolic activity. When cells are metabolically active they maintain a reducing environment

within the cytosol [128]. Resazurin is the active molecule of AlamarBlue® reagent, blue in color,

non-toxic and cell permeable. Upon within the cells, resazurin is reduced to resorufin, turning

from blue to red and highly fluorescent. Viable cells continuously convert resazurin to resorufin,

increasing the overall fluorescence and color of the media surrounding the cells, that can be

measured [129].

In Chapter VIII, Alamar blue was performed to assess the effect of CMCht/PAMAM, YIGSR-

CMCht/PAMAM and GEM-loaded YIGSR-CMCht/PAMAM dendrimer NPs in terms of metabolic

activity in each cell type separately (HCT-115 and L929). Cells were grown in Dulbecco’s

Chapter IV – Materials and Methods

112

Modified Eagle Medium (DMEM, Gibco, Invitrogen) supplemented with 10 % (v/v) fetal bovine

serum and 1 % (v/v) penicillin/streptomycin at 37 °C and 5 % CO2. Subcultures of cells were

performed when confluence reached values of ~90 %. Then, cells were plated at an initial density

of 1x104 cells per well in a 24-well plate. The working solution consists of 20 % (v/v)

AlamarBlue® (BioRad, Hercules, CA, USA) solution and 80 % culture medium, which was

prepared and protected from light (Alamar Blue working solution). At the end of each time-point,

cells were washed with PBS solution and immersed in 500 µL of Alamar Blue working solution.

The culture plates were kept in the dark, at 37 ºC in the CO2 incubator for 4 hours. The reacted

AlamarBlue was read in a microplate reader (Synergy HT, BioTek, Instruments, USA) using 100

µL supernatant per well, at an excitation wavelength of 530/25 nm and at an emission

wavelength of 590/35 nm. Three independent experiments were performed. The metabolic

activity values were calculated by normalization with the mean fluorescence value obtained for

the controls (Tissue culture polystyrene; TCPs) with and without cells.

IV-8.5.2. ATP bioluminescence assay

In Chapter VI, a CellTiter-Glo® Luminescent Cell Viability Assay (Promega, WI, USA) was

used to evaluate the viability of the HCT-116 that had migrated towards the center of the SF

hydrogels in the response to the chemoattractant hVCAM-1 (and without). This is a homogeneous

and reproducible method for the determination of cell viability based on the quantification of the

existing Adenosine Triphospate (ATP), which directly quantifies the metabolic activity of cells. This

assay was designed for direct use in multi-well culture plates, making it ideal for an automated

screening of cells cytotoxicity, and highly reproducible. This assay involves adding a single

reagent (CellTiter-Glo® Reagent) directly to the cultured cells/hydrogels, prepared in the

corresponding serum-free culture medium. The contact with this reagent induces cell lysis and

generation of a luminescent signal proportional to the amount of the present ATP. The amount of

ATP is directly proportional to the number of cells in culture [130].

In Chapter VI, the cell-laden hydrogels were retrieved from Vena4™ chips after each

timepoint and placed in 24-well plates; washed with PBS solution and incubated in 1 mL of ATP

working solution, consisting of CellTiter-Glo® Reagent and serum-free cell culture medium DMEM

(without phenol red), in a 1:1 ratio. The specimens were incubated for 30 minutes at RT, in an

Chapter IV – Materials and Methods

113

orbital shaker (KS 260 control, Wilmington, NC, USA). The reacted ATP solution was read in the

microplate reader using 100 µL supernatant per well, and the emitted luminescence was

detected using a sensitivity of 120. Different concentrations of ATP disodium salt solution,

ranging between 0 and 2 µmol/L, were used to prepare a standard curve for the calculation of

ATP concentration, relating quantity of ATP and luminescence intensity. Three specimens were

tested at each time-point. Three independent experiments were performed.

IV-8.5.3. DNA Quantification

DNA content was analyzed by the total double stranded DNA (dsDNA) by using the Pico-

Green® dsDNA quantification assay. This colorimetric assay employs fluorescence to evaluate

cell proliferation through the measurement of the dsDNA of samples. When the PicoGreen®

fluorescent marker is added to the solution, it specifically binds to the dsDNA emitting

fluorescence, read at 480 nm (excitation) and 520 nm (emission). DNA content is further

determined through a DNA standard curve, prepared using standard dsDNA solutions with

concentrations ranging from 0 to 2 μg/mL.

In Chapter V, after the determined timepoints, HeLa, GBM and HCT-116 cancer cells were

washed with sterile PBS and were lysed with 1 mL of ultrapure water in each well. The cellular

suspensions were transferred into microtubes and incubated in a water bath at 37 °C for 1 hour,

then stored at -80 °C for further analysis (at least 6 hours before the DNA quantification assay).

In Chapter VIII DNA quantification was performed to assess the effect of CMCht/PAMAM,

YIGSR-CMCht/PAMAM and GEM loaded-YIGSR-CMCht/PAMAM dendrimer NPs on cancer cells

and fibroblasts separately. After the determined timepoints, cells seeded were washed with sterile

PBS and lysed with 1 mL of ultrapure water in each well. The cellular suspensions were

transferred into microtubes, incubated in a water bath at 37 °C for 1 hour, and stored at -80 °C

for further analysis.

For the DNA quantification, the samples were defrosted at RT, and then underwent

ultrasonication treatment for 1 hour at 37 ºC to release the DNA. The doublestranded DNA

(dsDNA) was quantified by using the Quant-iT™ Pico-Green® dsDNA kit 2000 assays (Thermo

Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions. Briefly, 28.7

μL of sample or standard were added to a 96-well white opaque plate, mixed with 100 μL of 1X

Chapter IV – Materials and Methods

114

Tris-EDTA buffer (10 mmol/L Tris-HCl, 1 mmol/L EDTA, pH 7.5) and 71.3 μL of 1X Quant-iT™

PicoGreen® reagent, in a total volume of 200 μL. The plate was incubated at RT for 10 minutes

and read in the microplate reader (ex/em 485/528). Three specimens of each group were tested

at each time-point. Three independent experiments were performed.

IV-8.5.4. MTS

The MTS assay is a colorimetric method commonly used for determining the number of

viable cells in proliferation or cytotoxicity assays. In this protocol, MTS assay is designed to

assess the short-term cytotoxicity of polymeric biomaterials, in this case NPs developed along this

thesis. The MTS assay is based on bio-reduction of a tetrazolium compound, 3-(4,5-

dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfofenyl)-2H-tetrazolium (MTS), into a

brown formazan product that is soluble in tissue culture medium. This conversion is presumably

accomplished by NADPH or NADH produced by dehydrogenase enzymes in metabolically active

cells. The quantity of formazan product, as measured by the amount of 490 nm absorbance, is

directly proportional to the number of living cells in culture

In Chapter V, MTS viability test was performed to determine the cytotoxicity of

CMCht/PAMAM dendrimer NPs in HeLa, U87MG and HCT-116 cancer cells in static and

dynamic conditions. Cells were continuously grown in DMEM (Gibco, MD, USA; Invitrogen, CA,

USA) supplemented with 10 % (v/v) fetal bovine serum and 1% (v/v) penicillin and streptomycin

under standard conditions (37 °C in a humidified atmosphere containing 5 % CO2). Subcultures

of cells were performed when confluence reached values of approximately 90 %. Then, cells were

plated at an initial density of 2x104 cells/well in 24-well plates and exposed to three

concentrations of CMCht/PAMAM dendrimer NPs (0.1, 0.5 and 1 mg/mL) for short-term

exposures comprised between 1 and 3 days. By the end of the timepoints, cells cultures (n=3)

were placed in culture medium containing MTS in a 5:1 ratio and incubated in a humidified

atmosphere at 37 ºC and 5 % CO2. After three hours of incubation, 100 µL of solution from each

well were transferred to a 96-well TCPS plate and the optical density was determined at 490 nm.

Cultures that were not exposed to the CMCht/PAMAM dendrimer NPs were used as controls. In

the case of the Vena8™ chip, to test cell viability using the MTS method, cells from each channel

were detached using Tryple Express (Alfagene, Portugal). Cells were aspirated and collected in a

Chapter IV – Materials and Methods

115

1.5 mL tube. After centrifugation (5 minutes, 1500G), cells were incubated with culture medium

containing MTS in a 5:1 ratio and analyzed as aforementioned for 2D standard cell cultures.

IV-8.5.5. Flow cytometry analysis

Flow cytometry is a sophisticated instrument that can measure multiple physical

characteristics of a single cell such as size and granularity simultaneously, as the cell flows in

suspension through a measuring device. Its working depends on the light scattering features of

the cells under investigation, which may be derived from dyes or monoclonal antibodies targeting

either extracellular molecules located on the surface or intracellular molecules inside the cells.

This approach makes flow cytometry a powerful tool for detailed analysis of complex populations

in a short period of time [131].

In Chapter V, flow cytometry was used as a quantitative method for determining cell viability

levels of labeled CMCht/PAMAM dendrimer nanoparticles inside of GBM, HeLa and HCT-116

cells under static and dynamic conditions. In static conditions, cancer cells were cultured in the

presence of 0.5/mg mL FITC-labeled CMCht/PAMAM dendrimer NPs in a 6-well plate (2x105

cells/well). Cells cultured in complete culture medium were used as controls. Cells were released

from a substratum and a cell strainer was used to avoid cell clusters. Afterward, 0.5 mL of

DMEM complete medium was added to each well and samples transferred to cytometry tubes. In

dynamic conditions, cancer cells were cultured in the presence of 0.5 mg/mL FITC-labeled

CMCht/PAMAM dendrimer NPs in the microfluidic chips. Cells cultured in complete culture

medium were used as controls. Flow cytometry analysis was carried out to perform quantitative

analysis on the internalization efficiency as well as cell viability, as follows: each channel was

washed with PBS. Then, PBS was aspirated, and cells from each channel were detached using

10 µL of TrypleExpress and collected in an FACS tube. After centrifugation at 1500×r.p.m. for 5

minutes, cells were re-suspended. in 800 µL of a 2 % (v/v) fetal bovine serum in PBS solution.

Afterward, 5 µL of 7-Aminoactinomycin D (7-AAD; Taper group, Portugal) was added to each

sample for determining the number of dead cells. After this step, cells were loaded in

FACSCalibur flow cytometer (BD Biosciences Immunocytometry Systems, CA, USA). Calibrate

beads three-color kit (BD CaliBRITE™ beads, CA, USA) was used to adjust the equipment

Chapter IV – Materials and Methods

116

instrument settings before samples are run on the flow cytometer. Finally, data were treated

using the Flowing Software 2.

In Chapter VIII, flow cytometry was used to quantify the number of dead cells in a co-culture

of HCT-116 and L929 cells cultured in DMEM supplemented with to 0.5 mg/mL of GEM loaded

YIGSR-CMCht/PAMAM. Cells were grown in Dulbecco’s Modified Eagle Medium (DMEM, Gibco,

Invitrogen) supplemented with 10 % (v/v) fetal bovine serum and 1 % (v/v)

penicillin/streptomycin at 37 °C and 5 % CO2. Subcultures of cells were performed when

confluence reached values of ~90 %. Then, co-culture was achieved by plating an initial density

of 1x104 of each cell type per well in a 24-well plate. Cells were cultured in medium

supplemented with 0.5 mg/mL of GEM-loaded YIGSR-CMCht/PAMAM. After 24 and 72 hours,

cells were trypsinized (TrypleX) and a cell strainer was used to avoid cell clusters. Next, 0.5 mL of

DMEM complete medium was added to each well and samples transferred to cytometry tubes.

To distinguish cells in co-culture, 10 µL of FITC-CD44 antibody (Abcam, Portugal, Ref. ab27285)

was incubated with co-cultured cells for 30 minutes at RT. After washing with 2 mL of 2 % FBS

(v/v) in PBS, the tubes were centrifuged at 300 rpm for 5 minutes and re-suspended in 500 µL

of % FBS (v/v) in PBS. Cell suspensions were analyzed in FACSCalibur flow cytometer after

adding 1 µL of Propidium iodide (PI). For the flow cytometry cell viability assays in 3D, cells

encapsulated in Matrigel were retrieved using 500 µL of Corning™ Cell Recovery Solution for 30

minutes at 4 ºC and the subsequent protocol was performed as previously described. For all

assays, calibrate beads three-color kit (BD CaliBRITETM beads, USA) was used to adjust the

equipment instrument settings before samples are run on the flow cytometer. Finally, data was

analyzed using the FLOWING SOFTWARE 2.

IV-8.5.6. Live/Dead staining assay

Calcein-acetoxymethyl (Calcein-AM)/Propidium iodide (PI) staining was used. It is a method

for the visualization of cell viability and death. In this method, cells are incubated with Calcein-

AM, which is a non-fluorescent and cell-permeant derivative of calcein, transported through the

cellular membrane. When the intracellular esterases of living cells remove the acetomethoxy

group, this probe becomes green fluorescent. However, it stays non-fluorescent when the

acetoxymethyl ester is intact because of the non-active esterases of dead cells. The specific

Chapter IV – Materials and Methods

117

visualization of dead cells can be performed by incubation with fluorescent PI, which shows

enhanced fluorescence when binding with high affinity to DNA. Contrary to calcein-AM, PI does

not permeant the cellular membrane, it binds to the cytoplasmic DNA when the cellular

membrane of dead cells is disrupted [132]. Instead of PI, ethidium homodimer (EthD-1) can also

be used.

In Chapter VI, Live/dead staining was used to determine whereas cells that migrated from

the microchannels of Vena4™ into the SF and Matrigel hydrogels were viable. This would be a

good indicator that the hydrogels allow cell migration but also if they allow flow of nutrients

enough to keep the cells alive. For this, the cell-laden SF and Matrigel constructs were removed

from the Vena4™ chip and washed with PBS solution. Next, cells were stained with calcein-AM (1

μg/mL) and PI (2 μg/mL), followed by incubation for 15 minutes in the dark, at 37 ºC in the CO2

incubator. Samples are immediately observed under fluorescence microscopy (Calcein-AM ex/em

495/515 nm; PI ex/em 495/635 nm) in the transmitted and reflected light microscope. Images

were acquired using the Zen microscope processing software, connected to the digital camera

AxioCam MR3. A Z-stack function was used to combine images at different depths into one final

image.

In Chapter VII the viability of cells cultured in the microfluidic chip’s central chamber was

assessed in the presence or absence of GEM loaded dendrimer nanoparticles using the

Live/dead assay. For this, at endpoints, the chip side-channels were washed with PBS (Sigma),

and then a solution of Calcein AM (1 μM) and EthD-1 (6 μM) was perfused thought the lateral

channels and let to incubated for 40 minutes, to reach the cells in the center of the chamber.

After this, the lateral channels were washed with PBS. Microfluidic chips were then imaged with

fluorescence microscope using automated acquisition for Z-stack and multicolor-channel.

Acquired multi-frame multicolour images were combined either using Nikon AR or FIJI software

(open-source software ImageJ focused on biological image analysis) for operations of processing

such as rotation, crop, color balance and related assemblies, considering initial scales and

aspect ratios. High-resolution Z-stack images were processed for 3D reconstruction through

deconvolution and 3D projections made by NIS-Elements AR (Nikon) software.

In Chapter VIII, cell viability of HCT-116 and L929 in co-culture in the presence of GEM

loaded YIGSR-CMCht/PAMAM was also assessed by Live/Dead staining. HCT-116 and L929

Chapter IV – Materials and Methods

118

confluent cells were detached from the T150 cell culture flasks using trypsin (0.25 % (v/v)

trypsin–EDTA solution), centrifuged at 300 g for 5 minutes, diluted in a new cell suspension and

counted in a cell counter KOVA® Glasstic® slide 10 chamber with grids (Garden Grove, CA,

USA). L929 were pre-labeled with blue cell tracker CellTracker Blue CMAC (Alfagene, Portugal)

(1x10−6 M) and incubated at 37 °C for 30 minutes, whereas HCT-116 cells were left unstained.

At the end of the last time-point (72 hours), three samples were washed with PBS and incubated

in 1 µg/mL calcein-AM and 2 µg/mL PI prepared in DMEM low glucose supplemented with

sodium bicarbonate and 1 % (v/v) antibiotic/antimycotic solution for 10 minutes in the dark at 37

ºC in the 5 % CO2. After washing in PBS, samples were immediately analyzed in a confocal

microscopy (Calcein-AM in green: ex/em 495/515 nm; PI in red: ex/em 495/635 nm; Cell

Tracker Blue CMAC in blue: ex/em 353/466 nm) (Leica, SP8, Germany).

In Chapter IX, Live/dead assay was performed again. This time to prove HCT-116 cancer

cells encapsulated in the silk hydrogel microfluidic platform were viable, receiving enough oxygen

and nutrients, until transformation of silk proteins conformation from amorphous to β-sheet. For

this, at the end of each time point (24 hours and 3 days), three samples were washed with PBS

and incubated in 1 µg/mL calcein-AM and 2 µg/mL EthD-1 prepared in DMEM low glucose

supplemented with sodium bicarbonate and 1 % (v/v) antibiotic/antimycotic solution for 10

minutes in the dark at 37 ºC in the 5 % CO2.

IV-8.6. Assessment of FITC-labeled CMCht/PAMAM dendrimer nanoparticles internalization

IV-8.6.1. Fluorescence microscopy and fluorescence-activated cell sorting (FACS) analyses

In this thesis, the internalization of the FITC-labeled CMCht/PAMAM dendrimer

nanoparticles and FITC-YIGSR-CMCht/PAMAM dendrimer nanoparticles were investigated using

different cell types (Chapters V and VIII), both by fluorescence microscopy and flow cytometry

analyses.

Regarding fluorescence microscopy, it is a major tool with which to monitor cell physiology,

and in this case, the internalization of fluorescent labeled NPs. In Chapter V, to assess the

internalization but also possible cell morphological changes, fluorescence microscopy was used.

A cell suspension was prepared and seeded on tissue culture polystyrene coverslips (Sarstedt

Chapter IV – Materials and Methods

119

Inc., USA) in 24-well plates (2x104 cells/well) and cultured with 0.5 mg/mL FITC labeled

CMCht/PAMAM dendrimer NPs, in standard culture conditions for 1 and 3 days. Then cells were

fixed with 4 % (v/v) formalin (Sigma, Germany) and stained for F-actin filaments of the

cytoskeleton and nuclei with Texas Red-X phalloidin (Molecular Probes, Invitrogen, USA) and with

4,6-diamidino-2-phenylindole, dilactate (DAPI blue, Molecular Probes), respectively, following

supplier’s protocol. For dynamic cultures, cells were cultured as described previously in the

microfluidic platform and were fixed with 10 µL per channel of 4 % (v/v) formalin (Sigma,

Germany) for 10 minutes. Then, 10 µL per channel of DAPI and Texas Red-X Phalloidin were

perfused into the microchannels to stain the nuclei and the cytoskeleton F-actin filaments,

respectively, and washed with PBS after 30 minutes. Finally, cells were observed under the

fluorescence microscope (AxioImager Z1, Zeiss Inc., Oberkochen, Germany), in triplicates.

In Chapter V, flow cytometry was performed to quantify and compare the internalization

levels of FITC-labeled CMCht/PAMAM dendrimer NPs in static and dynamic cultures in three

cancer cell lines (GBM, HeLa and HCT-116). In static conditions, cancer cells were cultured in

the presence of 0.5 mg/mL FITC-labeled CMCht/PAMAM dendrimer NPs in a 6-well plate (2x105

cells/well). Cells cultured in complete culture medium were used as controls. Cells were released

from the substratum with Tryple Express and a cell strainer was used to avoid cell clusters.

Afterwards, 0.5 mL of DMEM complete medium was added to each well and samples transferred

to cytometry tubes, centrifuged and re-suspended in 800 µL of a 2 % (v/v) fetal bovine serum in

PBS solution. In dynamic conditions, cancer cells were cultured in the presence of 0.5 mg/mL

FITC-labeled CMCht/ PAMAM dendrimer NPs in the microfluidic chips. Cells cultured in complete

culture medium were used as controls. Each channel was washed with PBS. Then, PBS was

aspirated and cells from each channel were detached using 10 µL of Tryple Express and

collected in an FACS tube. After centrifugation at 1,500 g for 5 minutes, cells were re-suspended

in 800 µL of a 2 % (v/v) fetal bovine serum in PBS solution.

In Chapter VIII, when studying the possible targeted internalization analysis of the modified

YIGSR dendrimer nanoparticles, L929 and HCT-116 cells were co-cultured in medium

supplemented with 0.5 mg/mL of FITC-labeled YIGSR-CMCht/PAMAM dendrimer NPs in 24-well

plates (1x104 cells each type/well). Cells cultured in complete culture medium were used as

controls. After 24 and 72 hours, cells were trypsinized and a cell strainer was used to avoid cell

clusters. Next, 0.5 mL of DMEM complete medium was added to each well and samples

Chapter IV – Materials and Methods

120

transferred to cytometry tubes and centrifuged. To distinguish cells in co-culture, 10 µL of PE-

CD44 antibody (Abcam, Portugal, Ref. ab46793) was incubated with cell suspension for 30

minutes at room temperature in 2 % (v/v) FBS in PBS. CD44 binds only to HCT-116 cancer cells,

and not the L929 cells. After washing with 2 mL of 2 % (v/v) FBS in PBS, the tubes were

centrifuged at 300 g for 5 minutes and re-suspended in 1 % (v/v) formalin in PBS. Cell

suspensions were analyzed in FACSCalibur flow cytometer. Finally, data was analyzed using the

FLOWING SOFTWARE 2.

IV-8.7. Gene expression profiles

IV-8.7.1. RNA isolation and first-strand cDNA synthesis

The expression of mRNA for specific apoptosis and proliferation-related genes of interest was

measured by real-time PCR analysis. Extraction of high-quality RNA is a crucial step in gene

expression profiling, and previous studies have shown that using kits for RNA extraction allows to

obtain a higher purity and efficiency in the process [133].

In Chapter VII, different extraction kits were used for total RNA isolation. The great challenge

in this genetic analysis was thee retrieval of cells from Matrigel and from the central chamber of

the microfluidic chip. For this, after each culture time, Matrigel from the central chamber was

retrieved by punching out with a 5 mm sterile biopsy punch (WPI, The Netherlands) and placed

in 24 well plates. The PDMS was separated from the hydrogel. Samples were washed with PBS

solution and Matrigel was depolymerized by incubation with 1 mL of Cell Recovery Solution™ (BD

Biosciences, The Netherlands) for 30 minutes at 4 ºC. After complete release from the Matrigel,

cells were transferred to an Eppendorf tube and centrifuged to a pellet by centrifugation at 300 G

for 5 minutes. Cells were washed with cold PBS and centrifuged again. Total RNA was extracted

from cells recovered from the 3D cultures using Phasemaker™ Tubes (Invitrogen), following the

manufacturer’s instructions. The pellet was then immersed in 500 mL TRIZOL reagent and

stored at 80 ºC until further use. Next, the samples were transferred to Phasemaker™ Tubes and

incubated for 5 minutes to permit complete dissociation of the nucleoproteins complex. 0.2 mL

of chloroform per 1 mL of TRIzol™ Reagent used for lysis was added and shaken vigorously by

hand for 15 seconds. After a 15 minutes incubation at RT, samples were centrifuged for 5

Chapter IV – Materials and Methods

121

minutes at 12000 g at 4 °C. The mixture separates into a lower red phenol-chloroform, and

interphase, and a colorless upper aqueous phase. The Phasemaker™ Gel forms a barrier

between the upper and lower phases. Then, 0.5 mL of isopropanol was added to the aqueous

phase and incubated for 10 minutes, followed by a 10 minutes centrifugation at 4 ºC. The pellet

is re-suspended in 75 % ethanol. The supernatant is discarded with a pipette and the pellet left to

dry.

RNA quantification and purity were assessed using a NanoDrop ND-1000 spectrophotometer

(NanoDrop Technologies). Afterwards, samples were defrosted at RT and underwent

ultrasonication treatment for 15 minutes at 20 ºC to ensure complete lysis and collection of the

cells from constructs. Samples with a 260/280 ratio between 1.6 and 2.0 were used for first-

strand complementary DNA (cDNA) synthesis, performed with the iScript cDNA Synthesis kit

(Quanta Biosciences, MD, USA). The obtained cDNA was used as template for the amplification

of the target genes using the iQSYBR Green Supermix (Bio-Rad), according to manufacturer’s

instructions. Briefly, 100 ng of the total RNA of each sample were used to generate single-

stranded cDNA by random priming with iScript Reverse Transcriptase (RT). A reaction mixture

consisting of 4 µL qScript Reaction Mix, 1 µL iScript Reverse Transcriptase (RT), RNA template

(100 ng) and nuclease-free water was prepared in a final volume of 20 µL. The single-strand

cDNA synthesis occurred by incubating the complete reaction mixture for 5 minutes at 22 ºC,

followed by 30 minutes at 42 ºC and terminated with an incubation of 5 minutes at 85 ºC.

IV-8.7.2. Real-time polymerase chain reaction (Real-time PCR)

Real time polymerase chain reaction (PCR) was used in Chapter VIII, as technique for

detecting and measuring PCR products of interest [134]. For starting the RNA is reverse-

transcripted into a single-strand cDNA chain using a reverse transcriptase polymerase chain

reaction (RT-PCR), as described in the section above. Afterwards, the oligonucleotide primers

(specific for the target genes) are allowed to anneal to the template cDNA, and with a Taq

polymerase starting reaction, a double-strand cDNA is produced. The three steps process of

denaturation, primer annealing and extension are repeated in at least 40 cycles to yield PCR

products, detected in real-time as the reaction progresses. The real-time detection of PCR

products is made by including in the reaction a fluorescent molecule that reports an increase in

Chapter IV – Materials and Methods

122

the amount of DNA. The measured fluorescence reflects the amount of amplified product in each

reaction cycle. The amplification of the target cDNA for real-time PCR quantification was

performed using the PerfeCta SYBR Green FastMix kit (Quanta Biosciences, Gaithersberg, MD,

USA), according to the manufacturer’s instructions. Briefly, 1 µL of single-strand cDNA products,

1 µmol/L of primer forward and reverse, 10 µL PerfeCta SYBR Green FastMix and nuclease-free

water, in a final volume of 20 µL, were prepared as reaction mixture. The primers sequence,

specific for each gene of interest, are described in Table IV-2. Each real-time PCR run was carried

out with an initial incubation at 95 ºC for 2 minutes, followed by 45 cycles of denaturation (95

ºC, 10 seconds), annealing (specific for each gene, 25 seconds) and extension (72 ºC, 30

seconds) in a Mastercycle® ep realplex real-time PCR system (Eppendorf, Hamburg, Germany).

The transcripts expression data were normalized to the endogenous housekeeping gene

Ubiquitin-40S ribosomal protein S27a (RPS27A) and the relative quantification was calculated

using as calibrators the results obtained for each target gene at day 1 in basal culture conditions,

according to the Livak (2-∆∆CT) method [135]. Briefly, the ΔCt values were obtained by the

difference between the Ct values of the target genes and the GAPDH gene. These values were

then normalized by subtracting the ΔCt value of the calibrator samples, in order to obtain the

ΔΔCt values.

Table IV-2 - Primers list for the studied genes.

Gene Forward Primer Reverse Primer

RPS27A GCTTGCCAGCAAAGATCAGT GAGGTTGAACCCTCGGATAC

Caspase - 3 GAAATTGTGGAATTGATGCGTGA CTACAACGATCCCCTCTGAAAAA

MMP-1 GGGGCTTTGATGTACCCTAGC TGTCACACGCTTTTGGGGTTT

KI67 GCTGGCTCCTGTTCACGTA CTGGGCTACACTGAGCACC

Chapter IV – Materials and Methods

123

IV-8.8. Immunocytochemistry

Immunocytochemistry is a detection method used for the visualization of specific antigens in

cells, detected by antigen-antibody binding [136]. The principle of this technique consists in the

use of specific fluorescent-labeled secondary antibodies that bind to the primary antibodies,

visualized under a fluorescence microscope. This procedure involves some preliminary steps

after cell culturing, including cell fixation, cell membrane permeabilization, which enables the

antibodies to enter into the cells when searching for intracellular antigens, and a blocking step

that blocks unspecific bindings of antibodies. These steps are intercalated with washing steps to

remove unbound antibodies. All the antibodies used in this thesis can be found in Table IV – 3.

In Chapter VII, this technique was applied after each timepoint. The chip was washed with

PBS through the lateral channels, and then cells were fixed with 4 % (v/v) paraformaldehyde in

PBS for 30 minutes at (RT), and washed again with PBS. Permeabilization of cells was done with

0.1 % (v/v) Triton X-100 (Sigma-Aldrich) for 15 minutes at RT, followed by washing and

incubation with blocking solution CAS-Block (Thermo Fisher Scientific) for 30 minutes at 37 °C.

Primary antibody for Ki-67 diluted in CAS Block (rabbit, ab833 R&D Systems, 1:100) was

subsequently incubated overnight at 4 °C. Samples were washed with PBS and incubated with

the secondary antibody Alexa Fluor ® 488 conjugated (1:500, anti-rabbit, R&D Systems) together

with high-affinity filamentous 698 actin probe phalloidin conjugated with Alexa Fluor ® 594

(Invitrogen™, 1:80) in CASblock for 3 hours at RT. The samples were then washed with PBS,

incubated with 4’,6-700 diamidino-2-phenylindole (Invitrogen™, DAPI, 1:100 in PBS) for 10

minutes at RT, and finally washed twice with PBS. The samples were kept in PBS in the dark at 4

°C until image analysis. High magnification images were acquired with Nikon Eclipse Ti-E, using

703 automated acquisition for Z-stack and multicolor-channel.

In Chapter VIII, HCT-116 and L929 cells in co-culture were fixed with 4 % (v/v) formalin for

10 minutes and washed with PBS. Cells were permeabilized with 0.1 % (v/v) Triton X-100 in PBS

for 5 minutes and blocked with 3 % (v/v) BSA in PBS for 30 minutes. Immunolabeling was

performed using a Rabbit anti-human monoclonal antibody against 67 kDa LR (Abcam, 1:200) as

primary antibody, prepared in PBS 1 % (wt%.) BSA and incubated overnight at 4 ºC. The

coverslips were incubated for 1 hour at RT with the respective secondary fluorochrome-

conjugated antibodies, anti-rabbit/mouse IgG (Invitrogen, dilution 1:1000), with Phalloidin

Chapter IV – Materials and Methods

124

(1:200) for 30 minutes (F-actin filaments) (WVR, Portugal) and with 4,6-Diamidino-2-phenyindole,

dilactate (DAPI 1:1000) (WVR, Portugal) for nuclei staining for 15 minutes. Immunofluorescence

was observed under transmitted and fluorescence microscopy (Zeiss, Germany) (LR in green:

ex/em 488/517; Phalloidin in red: ex/em 594/618; DAPI in blue: ex/em 358/461).

Table IV-3 – Antibodies list used in immunocytochemistry.

Antibody Type Reference Host Dilution Reactivity

Ki-67 Primary ab833 Rabbit 1:100 Human,

Mouse, Rat

67kDa Laminin receptor Primary ab133645 Rabbit 1:500 Human,

Mouse, Rat

Antibody to rabbit IgG

(Alexa Fluor 488)

Secondary IC1051G

Donkey 1:500 Human

Antibody to rabbit IgG

(Alex Fluor 488)

Secondary A21206 Donkey 1:1000 Human

IV-9. STATISTICAL ANALYSIS

All quantitative data are presented as mean ± standard deviation (SD). Statistical analysis

was performed using Graph Pad Prism 6.00 software (San Diego, USA). Statistical significances

(∗p ≤ 0.05, ∗∗p ≤ 0.01 and ∗∗∗p ≤ 0.001) were determined using specific statistical tests

described in the subsection of materials and methods in the different chapters.

In Chapter V, all the performed tests were conducted in triplicate and the obtained data

points are presented as mean ± standard deviation (SD). The nonparametric Mann–Whitney test

was used to compare two groups and the Tukey’s test was used to compare results of

dendrimers’ cytotoxicity after normalization. Statistical significance was defined as p<0.05 for a

95% CI. All measurements were taken in triplicate.

In Chapter VI, the nonparametric Mann–Whitney test was used to compare two groups,

whereas comparison between more than two groups was performed using the Kruskal–Wallis

test followed by Dunn’s comparison test. A value of p<0.05 was considered statistically

significant. All measurements were taken in triplicate.

Chapter IV – Materials and Methods

125

In Chapter VII, the nonparametric Mann-Whitney test was used to compare two groups. A

value of p<0.05 was considered statistically significant. For multiple comparisons, the parametric

test one-way analysis of variance (ANOVA) was used, with P<0.05 considered statistically

significant. All measurements were taken in triplicate.

In Chapter VIII, the non-parametric Mann–Whitney test was used to compare two groups,

whereas comparison between more than two groups was performed using the Kruskal–Wallis

test followed by Dunn’s comparison test. A value of p<0.05 was considered statistically

significant. All measurements were taken in triplicate.

IV-10. REFERENCES

1. O'Brien, F.J., Biomaterials & scaffolds for tissue engineering. Materials Today, 2011. 14(3): p. 88-95. 2. Tolesa, L.D., B.S. Gupta, and M.-J. Lee, Chitin and chitosan production from shrimp shells using ammonium-based ionic liquids. International Journal of Biological Macromolecules, 2019. 130: p. 818-826. 3. Philibert, T., B.H. Lee, and N. Fabien, Current Status and New Perspectives on Chitin and Chitosan as Functional Biopolymers. Appl Biochem Biotechnol, 2017. 181(4): p. 1314-1337. 4. Fiamingo, A. and S.P. Campana-Filho, Structure, morphology and properties of genipin-crosslinked carboxymethylchitosan porous membranes. Carbohydr Polym, 2016. 143: p. 155-63. 5. Khattak, S., et al., Applications of cellulose and chitin/chitosan derivatives and composites as antibacterial materials: current state and perspectives. Applied Microbiology and Biotechnology, 2019. 103(5): p. 1989-2006. 6. de Abreu, F.R. and S.P. Campana-Filho, Characteristics and properties of carboxymethylchitosan. Carbohydrate Polymers, 2009. 75(2): p. 214-221. 7. Song, Q., et al., Synthesis and property studies of N-carboxymethyl chitosan. Journal of Applied Polymer Science, 2011. 119(6): p. 3282-3285. 8. Zheng, M., et al., Synthesis, characterization and biological safety of O-carboxymethyl chitosan used to treat Sarcoma 180 tumor. Carbohydrate Polymers, 2011. 86(1): p. 231-238. 9. Ren, C., D. Zhao, and L. Zhu, Use of N,O-carboxymethyl chitosan to prevent postsurgical adhesions in a rabbit double uterine horn model: a randomized controlled design. Sci China Life Sci, 2016. 59(5): p. 504-9.

Chapter IV – Materials and Methods

126

10. Upadhyaya, L., et al., The implications of recent advances in carboxymethyl chitosan based targeted drug delivery and tissue engineering applications. Journal of Controlled Release, 2014. 186: p. 54-87. 11. Chen, X.-G. and H.-J. Park, Chemical characteristics of O-carboxymethyl chitosans related to the preparation conditions. Carbohydrate Polymers, 2003. 53(4): p. 355-359. 12. Kundu, B., et al., Silk fibroin biomaterials for tissue regenerations. Advanced Drug Delivery Reviews, 2013. 65(4): p. 457-470. 13. Yang, M., et al., Preparation of porous scaffolds from silk fibroin extracted from the silk gland of Bombyx mori (B. mori). Int J Mol Sci, 2012. 13(6): p. 7762-75. 14. Dinjaski, N. and D.L. Kaplan, Recombinant protein blends: silk beyond natural design. Curr Opin Biotechnol, 2016. 39: p. 1-7. 15. Fu, C., Z. Shao, and V. Fritz, Animal silks: their structures, properties and artificial production. Chemical Communications, 2009(43): p. 6515-6529. 16. Omenetto, F.G. and D.L. Kaplan, New opportunities for an ancient material. Science, 2010. 329(5991): p. 528-31. 17. Ribeiro, V.P., et al., Rapidly responsive silk fibroin hydrogels as an artificial matrix for the programmed tumor cells death. PLOS ONE, 2018. 13(4): p. e0194441. 18. Vollrath, F., D. Porter, and C. Dicko, 5 - The structure of silk, in Handbook of Textile Fibre Structure, S.J. Eichhorn, et al., Editors. 2009, Woodhead Publishing. p. 146-198. 19. Koh, L.-D., et al., Structures, mechanical properties and applications of silk fibroin materials. Progress in Polymer Science, 2015. 46: p. 86-110. 20. Altman, G.H., et al., Silk-based biomaterials. Biomaterials, 2003. 24(3): p. 401-16. 21. Tao, H., et al., Metamaterial silk composites at terahertz frequencies. Adv Mater, 2010. 22(32): p. 3527-31. 22. Castrillón Martínez, D.C., et al., Characterization of sericin obtained from cocoons and silk yarns. Procedia Engineering, 2017. 200: p. 377-383. 23. Qi, Y., et al., A Review of Structure Construction of Silk Fibroin Biomaterials from Single Structures to Multi-Level Structures. Int J Mol Sci, 2017. 18(3). 24. Zeng, C., et al., Silk fibroin porous scaffolds for nucleus pulposus tissue engineering. Mater Sci Eng C Mater Biol Appl, 2014. 37: p. 232-40. 25. Panda, D., et al., Synthesis and viscoelastic characterization of microstructurally aligned Silk fibroin sponges. J Mech Behav Biomed Mater, 2017. 71: p. 362-371. 26. Jackman, S.L., et al., Silk Fibroin Films Facilitate Single-Step Targeted Expression of Optogenetic Proteins. Cell Rep, 2018. 22(12): p. 3351-3361. 27. Carvalho, C.R., et al., Tunable Enzymatically Cross-Linked Silk Fibroin Tubular Conduits for Guided Tissue Regeneration. Adv Healthc Mater, 2018. 7(17): p. e1800186. 28. Fang, Y., L. Xu, and M. Wang, High-Throughput Preparation of Silk Fibroin Nanofibers by Modified Bubble-Electrospinning. Nanomaterials (Basel, Switzerland), 2018. 8(7): p. 471. 29. Zhang, X., et al., Silk fibroin microfibers and chitosan modified poly (glycerol sebacate) composite scaffolds for skin tissue engineering. Polymer Testing, 2017. 62: p. 88-95. 30. Ribeiro, V.P., et al., Rapidly responsive silk fibroin hydrogels as an artificial matrix for the programmed tumor cells death. PloS one, 2018. 13(4): p. e0194441-e0194441. 31. Ribeiro, V.P., et al., Silk-based anisotropical 3D biotextiles for bone regeneration. Biomaterials, 2017. 123: p. 92-106. 32. Abdel-Fattah, W.I., N. Atwa, and G.W. Ali, Influence of the protocol of fibroin extraction on the antibiotic activities of the constructed composites. Progress in biomaterials, 2015. 4(2-4): p. 77-88.

Chapter IV – Materials and Methods

127

33. Chen, X., et al., Regenerated Bombyx silk solutions studied with rheometry and FTIR. Polymer, 2001. 42(25): p. 09969-09974. 34. Pawcenis, D., et al., Size exclusion chromatography for analyses of fibroin in silk: optimization of sampling and separation conditions. Applied Physics A, 2014. 114(2): p. 301-308. 35. Karageorgiou, V., et al., Porous silk fibroin 3-D scaffolds for delivery of bone morphogenetic protein-2 in vitro and in vivo. Journal of Biomedical Materials Research Part A, 2006. 78A(2): p. 324-334. 36. Hughes, C.S., L.M. Postovit, and G.A. Lajoie, Matrigel: a complex protein mixture required for optimal growth of cell culture. Proteomics, 2010. 10(9): p. 1886-90. 37. Orkin, R.W., et al., A murine tumor producing a matrix of basement membrane. The Journal of experimental medicine, 1977. 145(1): p. 204-220. 38. Kleinman, H.K. and G.R. Martin, Matrigel: basement membrane matrix with biological activity. Semin Cancer Biol, 2005. 15(5): p. 378-86. 39. Hughes, C.S., L.M. Postovit, and G.A. Lajoie, Matrigel: A complex protein mixture required for optimal growth of cell culture. PROTEOMICS, 2010. 10(9): p. 1886-1890. 40. Kleinman, H.K., R.J. Klebe, and G.R. Martin, Role of collagenous matrices in the adhesion and growth of cells. The Journal of cell biology, 1981. 88(3): p. 473-485. 41. Erickson, A.C. and J.R. Couchman, Still more complexity in mammalian basement membranes. J Histochem Cytochem, 2000. 48(10): p. 1291-306. 42. Tibbitt, M.W. and K.S. Anseth, Hydrogels as extracellular matrix mimics for 3D cell culture. Biotechnology and bioengineering, 2009. 103(4): p. 655-663. 43. Frantz, C., K.M. Stewart, and V.M. Weaver, The extracellular matrix at a glance. Journal of cell science, 2010. 123(Pt 24): p. 4195-4200. 44. Justus, C.R., et al., In vitro cell migration and invasion assays. Journal of visualized experiments : JoVE, 2014(88): p. 51046. 45. Yin, G.N., et al., Matrigel-based sprouting endothelial cell culture system from mouse corpus cavernosum is potentially useful for the study of endothelial and erectile dysfunction related to high-glucose exposure. J Sex Med, 2012. 9(7): p. 1760-72. 46. DeCicco-Skinner, K.L., et al., Endothelial cell tube formation assay for the in vitro study of angiogenesis. Journal of visualized experiments : JoVE, 2014(91): p. e51312-e51312. 47. Uemura, M., et al., Matrigel supports survival and neuronal differentiation of grafted embryonic stem cell-derived neural precursor cells. J Neurosci Res, 2010. 88(3): p. 542-51. 48. Tahergorabi, Z. and M. Khazaei, A review on angiogenesis and its assays. Iranian journal of basic medical sciences, 2012. 15(6): p. 1110-1126. 49. Li, L., et al., How to Improve the Survival of Transplanted Mesenchymal Stem Cell in Ischemic Heart? Stem cells international, 2016. 2016: p. 9682757-9682757. 50. Albini, A., Extracellular Matrix Invasion in Metastases and Angiogenesis: Commentary on the Matrigel “Chemoinvasion Assay”. Cancer Research, 2016. 76(16): p. 4595. 51. Anguiano, M., et al., Characterization of three-dimensional cancer cell migration in mixed collagen-Matrigel scaffolds using microfluidics and image analysis. PloS one, 2017. 12(2): p. e0171417-e0171417. 52. Wolf, M.P., G.B. Salieb-Beugelaar, and P. Hunziker, PDMS with designer functionalities—Properties, modifications strategies, and applications. Progress in Polymer Science, 2018. 83: p. 97-134. 53. Dinh, T.-H.-N., et al., Mechanical Characterization of PDMS Films for the Optimization of Polymer Based Flexible Capacitive Pressure Microsensors. Journal of Sensors, 2017. 2017: p. 9.

Chapter IV – Materials and Methods

128

54. Larson, B.J., et al., Long-term reduction in poly(dimethylsiloxane) surface hydrophobicity via cold-plasma treatments. Langmuir, 2013. 29(42): p. 12990-6. 55. Xu, B.-Y., et al., One step high quality poly(dimethylsiloxane)-hydrocarbon plastics bonding. Biomicrofluidics, 2012. 6(1): p. 16507-165078. 56. Kim, B., K.P. ET, and I. Papautsky, Long-term stability of plasma oxidized PDMS surfaces. Conf Proc IEEE Eng Med Biol Soc, 2004. 7: p. 5013-6. 57. Piruska, A., et al., The autofluorescence of plastic materials and chips measured under laser irradiation. Lab Chip, 2005. 5(12): p. 1348-54. 58. Peterson, S.L., et al., Poly(dimethylsiloxane) thin films as biocompatible coatings for microfluidic devices: cell culture and flow studies with glial cells. J Biomed Mater Res A, 2005. 72(1): p. 10-8. 59. Zhou, J., A.V. Ellis, and N.H. Voelcker, Recent developments in PDMS surface modification for microfluidic devices. Electrophoresis, 2010. 31(1): p. 2-16. 60. Kikkawa, Y., et al., Laminin-111-derived peptides and cancer. Cell adhesion & migration, 2013. 7(1): p. 150-256. 61. Fridman, R., et al., Enhanced tumor growth of both primary and established human and murine tumor cells in athymic mice after coinjection with Matrigel. J Natl Cancer Inst, 1991. 83(11): p. 769-74. 62. Kawasaki, K., et al., Amino acids and peptides. XIV. Laminin related peptides and their inhibitory effect on experimental metastasis formation. Biochem Biophys Res Commun, 1991. 174(3): p. 1159-62. 63. Yoshida, N., et al., The laminin-derived peptide YIGSR (Tyr-Ile-Gly-Ser-Arg) inhibits human pre-B leukaemic cell growth and dissemination to organs in SCID mice. British journal of cancer, 1999. 80(12): p. 1898-1904. 64. Khalfaoui, T., et al., Laminin receptor 37/67LR regulates adhesion and proliferation of normal human intestinal epithelial cells. PLoS One, 2013. 8(8): p. e74337. 65. Kumazoe, M., et al., 67-kDa laminin receptor increases cGMP to induce cancer-selective apoptosis. J Clin Invest, 2013. 123(2): p. 787-99. 66. Vacca, A., et al., Melanocyte tumor progression is associated with changes in angiogenesis and expression of the 67-kilodalton laminin receptor. Cancer, 1993. 72(2): p. 455-61. 67. Shi, Y.E., et al., Expression of 67 kDa laminin receptor in human breast cancer cells: regulation by progestins. Clin Exp Metastasis, 1993. 11(3): p. 251-61. 68. Vince, D.G., J.A. Hunt, and D.F. Williams, Quantitative assessment of the tissue response to implanted biomaterials. Biomaterials, 1991. 12(8): p. 731-6. 69. Nerem, R.M. and S.C. Schutte, Chapter 2 - The Challenge of Imitating Nature, in Principles of Tissue Engineering (Fourth Edition), R. Lanza, R. Langer, and J. Vacanti, Editors. 2014, Academic Press: Boston. p. 9-24. 70. Meek, M.F., et al., In vitro degradation and biocompatibility of poly(DL-lactide-epsilon-caprolactone) nerve guides. J Biomed Mater Res A, 2004. 68(1): p. 43-51. 71. Abbasi, E., et al., Dendrimers: synthesis, applications, and properties. Nanoscale research letters, 2014. 9(1): p. 247-247. 72. Tomalia, D.A. and J.M.J. Fréchet, Discovery of dendrimers and dendritic polymers: A brief historical perspective*. Journal of Polymer Science Part A: Polymer Chemistry, 2002. 40(16): p. 2719-2728.

Chapter IV – Materials and Methods

129

73. Palmerston Mendes, L., J. Pan, and V.P. Torchilin, Dendrimers as Nanocarriers for Nucleic Acid and Drug Delivery in Cancer Therapy. Molecules (Basel, Switzerland), 2017. 22(9): p. 1401. 74. Oliveira, J.M., et al., Dendrimers and derivatives as a potential therapeutic tool in regenerative medicine strategies—A review. Progress in Polymer Science, 2010. 35(9): p. 1163-1194. 75. Oliveira, J.M., et al., The osteogenic differentiation of rat bone marrow stromal cells cultured with dexamethasone-loaded carboxymethylchitosan/poly(amidoamine) dendrimer nanoparticles. Biomaterials, 2009. 30(5): p. 804-13. 76. Xu, L., H. Zhang, and Y. Wu, Dendrimer advances for the central nervous system delivery of therapeutics. ACS Chem Neurosci, 2014. 5(1): p. 2-13. 77. Dufes, C., I.F. Uchegbu, and A.G. Schatzlein, Dendrimers in gene delivery. Adv Drug Deliv Rev, 2005. 57(15): p. 2177-202. 78. Sk, U.H. and C. Kojima, Dendrimers for theranostic applications. Biomol Concepts, 2015. 6(3): p. 205-17. 79. Li, D.-Y., et al., On-surface synthesis of planar dendrimers via divergent cross-coupling reaction. Nature Communications, 2019. 10(1): p. 2414. 80. Hawker, C. and J.M.J. Fréchet, A new convergent approach to monodisperse dendritic macromolecules. Journal of the Chemical Society, Chemical Communications, 1990(15): p. 1010-1013. 81. Jevprasesphant, R., et al., Engineering of Dendrimer Surfaces to Enhance Transepithelial Transport and Reduce Cytotoxicity. Pharmaceutical Research, 2003. 20(10): p. 1543-1550. 82. Han, W., et al., Synthesis of novel poly(ester amine) dendrimers by Michael addition and acrylate esterification. Designed Monomers and Polymers, 2013. 16(1): p. 67-71. 83. Tam, R.Y., L.J. Smith, and M.S. Shoichet, Engineering Cellular Microenvironments with Photo- and Enzymatically Responsive Hydrogels: Toward Biomimetic 3D Cell Culture Models. 2017. 50(4): p. 703-713. 84. Li, Y. and E. Kumacheva, Hydrogel microenvironments for cancer spheroid growth and drug screening. Science advances, 2018. 4(4): p. eaas8998-eaas8998. 85. Geckil, H., et al., Engineering hydrogels as extracellular matrix mimics. Nanomedicine (London, England), 2010. 5(3): p. 469-484. 86. Hoffman, A.S., Hydrogels for biomedical applications. Advanced drug delivery reviews, 2012. 64: p. 18-23. 87. Singh, N.K. and D.S. Lee, In situ gelling pH- and temperature-sensitive biodegradable block copolymer hydrogels for drug delivery. Journal of Controlled Release, 2014. 193: p. 214-227. 88. Teixeira, L.S.M., et al., Enzyme-catalyzed crosslinkable hydrogels: emerging strategies for tissue engineering. Biomaterials, 2012. 33(5): p. 1281-1290. 89. Yan, L.P., et al., Core‐shell silk hydrogels with spatially tuned conformations as

drug‐delivery system. Journal of tissue engineering and regenerative medicine, 2016.

90. Yan, L.P., et al., Tumor Growth Suppression Induced by Biomimetic Silk Fibroin Hydrogels. Scientific Reports, 2016. 6: p. 31037. 91. Yucel, T., P. Cebe, and D.L. Kaplan, Vortex-induced injectable silk fibroin hydrogels. Biophysical journal, 2009. 97(7): p. 2044-2050. 92. Fini, M., et al., The healing of confined critical size cancellous defects in the presence of silk fibroin hydrogel. Biomaterials, 2005. 26(17): p. 3527-3536.

Chapter IV – Materials and Methods

130

93. Wang, X., et al., Sonication-induced gelation of silk fibroin for cell encapsulation. Biomaterials, 2008. 29(8): p. 1054-1064. 94. Yan, L.-P., et al., Macro/microporous silk fibroin scaffolds with potential for articular cartilage and meniscus tissue engineering applications. Acta biomaterialia, 2012. 8(1): p. 289-301. 95. Liu, B., et al., Silk structure and degradation. Colloids and Surfaces B: Biointerfaces, 2015. 131: p. 122-128. 96. Sofia, S., et al., Functionalized silk-based biomaterials for bone formation. Journal of Biomedical Materials Research, 2001. 54(1): p. 139-148. 97. Jin, H.-J. and D.L. Kaplan, Mechanism of silk processing in insects and spiders. Nature, 2003. 424(6952): p. 1057-1061. 98. Olanrewaju, A., et al., Capillary microfluidics in microchannels: from microfluidic networks to capillaric circuits. Lab Chip, 2018. 18(16): p. 2323-2347. 99. Huh, D., et al., Microfabrication of human organs-on-chips. Nat Protoc, 2013. 8(11): p. 2135-57. 100. King, M.C., Chapter 2 - Principles of Optical Lithography, in VLSI Electronics Microstructure Science, N.G. Einspruch, Editor. 1981, Elsevier. p. 41-81. 101. Brower, K., A.K. White, and P.M. Fordyce, Multi-step Variable Height Photolithography for Valved Multilayer Microfluidic Devices. J Vis Exp, 2017(119). 102. Liston, E.M., L. Martinu, and M.R. Wertheimer, Plasma surface modification of polymers for improved adhesion: a critical review. Journal of Adhesion Science and Technology, 1993. 7(10): p. 1091-1127. 103. Qin, D., Y. Xia, and G.M. Whitesides, Soft lithography for micro- and nanoscale patterning. Nature Protocols, 2010. 5: p. 491. 104. Maghsoudi, K., et al., Micro-nanostructured polymer surfaces using injection molding: A review. Materials Today Communications, 2017. 13: p. 126-143. 105. Pugieux, C., et al., Chapter Twenty-Four - Spindle Assembly on Immobilized Chromatin Micropatterns, in Methods in Enzymology, R.D. Vale, Editor. 2014, Academic Press. p. 435-448. 106. Al-Abboodi, A., et al. Microfluidic chip containing porous gradient for chemotaxis study. 2011. 107. Saadi, W., et al., A parallel-gradient microfluidic chamber for quantitative analysis of breast cancer cell chemotaxis. Biomed Microdevices, 2006. 8(2): p. 109-18. 108. Deneke, V.E. and S. Di Talia, Chemical waves in cell and developmental biology. The Journal of Cell Biology, 2018. 217(4): p. 1193. 109. Damiati, S., et al., Microfluidic Devices for Drug Delivery Systems and Drug Screening. Genes, 2018. 9(2): p. 103. 110. Weltin, A., et al., Cell culture monitoring for drug screening and cancer research: a transparent, microfluidic, multi-sensor microsystem. Lab Chip, 2014. 14(1): p. 138-46. 111. Hayes, T.L. and R.F.W. Pease, The Scanning Electron Microscope: Principles and Applications in Biology and Medicine, in Advances in Biological and Medical Physics, J.H. Lawrence and J.W. Gofman, Editors. 1968, Elsevier. p. 85-137. 112. Ismail, A.A., F.R. van de Voort, and J. Sedman, Chapter 4 Fourier transform infrared spectroscopy: Principles and applications, in Techniques and Instrumentation in Analytical Chemistry, J.R.J. Paré and J.M.R. Bélanger, Editors. 1997, Elsevier. p. 93-139. 113. Winey, M., et al., Conventional transmission electron microscopy. Mol Biol Cell, 2014. 25(3): p. 319-23.

Chapter IV – Materials and Methods

131

114. Marrese, M., V. Guarino, and L. Ambrosio, Atomic Force Microscopy: A Powerful Tool to Address Scaffold Design in Tissue Engineering. Journal of functional biomaterials, 2017. 8(1): p. 7. 115. A Practical Introduction to Differential Scanning Calorimetry, in Principles and Applications of Thermal Analysis. 116. Bhattacharjee, S., DLS and zeta potential – What they are and what they are not? Journal of Controlled Release, 2016. 235: p. 337-351. 117. Hepokur, C., et al., Silver nanoparticle/capecitabine for breast cancer cell treatment. Toxicology in Vitro, 2019. 61: p. 104600. 118. Nejadnik, M.R. and W. Jiskoot, Measurement of the average mass of proteins adsorbed to a nanoparticle by using a suspended microchannel resonator. J Pharm Sci, 2015. 104(2): p. 698-704. 119. Barnes, H.A., J.F. Hutton, and K. Walters, An introduction to rheology. Vol. 3. 1989: Elsevier. 120. Kaur, G. and J.M. Dufour, Cell lines: Valuable tools or useless artifacts. Spermatogenesis, 2012. 2(1): p. 1-5. 121. Entschladen, F., et al., Tumour-cell migration, invasion, and metastasis: navigation by neurotransmitters. Lancet Oncol, 2004. 5(4): p. 254-8. 122. Gupta, G.P. and J. Massagué, Cancer Metastasis: Building a Framework. Cell, 2006. 127(4): p. 679-695. 123. Boyden, S., The chemotactic effect of mixtures of antibody and antigen on polymorphonuclear leucocytes. J Exp Med, 1962. 115: p. 453-66. 124. Chen, H.C., Boyden chamber assay. Methods Mol Biol, 2005. 294: p. 15-22. 125. Nishida, N., et al., Angiogenesis in cancer. Vascular health and risk management, 2006. 2(3): p. 213-219. 126. Yeon, J.H., et al., In vitro formation and characterization of a perfusable three-dimensional tubular capillary network in microfluidic devices. Lab Chip, 2012. 12(16): p. 2815-22. 127. Weis, S., et al., Endothelial barrier disruption by VEGF-mediated Src activity potentiates tumor cell extravasation and metastasis. J Cell Biol, 2004. 167(2): p. 223-9. 128. Rampersad, S.N., Multiple applications of Alamar Blue as an indicator of metabolic function and cellular health in cell viability bioassays. Sensors (Basel, Switzerland), 2012. 12(9): p. 12347-12360. 129. O'brien, J., et al., Investigation of the Alamar Blue (resazurin) fluorescent dye for the assessment of mammalian cell cytotoxicity. The FEBS Journal, 2000. 267(17): p. 5421-5426. 130. Crouch, S., et al., The use of ATP bioluminescence as a measure of cell proliferation and cytotoxicity. Journal of immunological methods, 1993. 160(1): p. 81-88. 131. Adan, A., et al., Flow cytometry: basic principles and applications. Crit Rev Biotechnol, 2017. 37(2): p. 163-176. 132. Kaneshiro, E.S., et al., Reliability of calcein acetoxy methyl ester and ethidium homodimer or propidium iodide for viability assessment of microbes. Journal of Microbiological Methods, 1993. 17(1): p. 1-16. 133. Kim, J.-H., et al., Comparison of three different kits for extraction of high-quality RNA from frozen blood. Springerplus, 2014. 3(1): p. 76. 134. Arya, M., et al., Basic principles of real-time quantitative PCR. Expert review of molecular diagnostics, 2005. 5(2): p. 209-219.

Chapter IV – Materials and Methods

132

135. Livak, K.J. and T.D. Schmittgen, Analysis of relative gene expression data using real-time quantitative PCR and the 2− ΔΔCT method. methods, 2001. 25(4): p. 402-408.

136. Maity, B., D. Sheff, and R.A. Fisher, Immunostaining: detection of signaling protein location in tissues, cells and subcellular compartments. Methods Cell Biol, 2013. 113: p. 81-105.

133

SECTION 3

EXPERIMENTAL SECTION

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

134

Chapter V

A Semiautomated Microfluidic

Platform for Real-time

Investigation of Nanoparticles’

Cellular Uptake and Cancer

Cell’s Tracking

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

135

Chapter V

Chapter V - A Semiautomated Microfluidic Platform for Real

time Investigation of Nanoparticles’ Cellular Uptake and

Cancer Cell’s Tracking1

ABSTRACT

Aim: Develop a platform composed of labeled dendrimer nanoparticles (NPs) and a microfluidic

device for real-time monitoring of cancer cells fate. Materials & methods:

Carboxymethylchitosan/poly(amidoamine) dendrimer NPs were labeled with fluorescein-5(6)-

isothiocyanate and characterized using different physicochemical techniques. After, HeLa, HCT-

116 and U87MG were cultured in the presence of NPs, and cell viability and internalization

efficiency in static (standard culture) and dynamic (microfluidic culture) conditions were

investigated. Results: Cancer cells cultured with NPs in dynamic conditions were viable and

presented higher internalization levels as compared with static 2D cultures. Conclusion: This

work demonstrated that the proposed microfluidic-based platform allows real-time monitoring,

which upon more studies, namely, the assessment of an anticancer drug release effect could be

used for cancer theranostics.

Keywords: Cell tracking; Circulating tumor cells; Microfluidics; Nanoparticles; Theranostics.

1This chapter is based on the following publication:

M.R. Carvalho, F. R. Maia, J. Silva-Correia, B. M. Costa, R. L. Reis, J. M. Oliveira. “A semi-automated microfluidic platform for real-

time tracking of cancer cells and investigation of nanoparticles cellular uptake”. Nanomedicine, 2016, 2:581-596. doi:

10.2217/nnm-2016-0344.

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

136

V-1. GRAPHICAL ABSTRACT

V-2. INTRODUCTION

Cancer is a major cause of mortality, being directly responsible for 1.665.540 new cancer

cases and 585.720 cancer deaths in the United States in 2014 [1]. Its incidence is estimated to

increase up to 13 million by 2030, worldwide [2]. The metastasis process, where circulating

tumor cells (CTCs) migrate and colonize distant organs, is responsible for about 90 % of deaths

[3]. Thus, CTCs have been highly studied and new information has arisen regarding its fate and

subsequent host interaction, the importance of shear stress and real time visualization in

understanding these processes [4]. New strategies for cancer diagnosis and treatment have

become more sophisticated and high throughput processes have been applied due to forward-

thinking nanotechnology techniques, such as the use of nanoparticles (NP). In fact, NP can be

modified to target specific cells, as CTCs, and induce specific cancer cell death [5]. Additionally,

these nanometer size structures are very attractive for health-related applications, as they can be

rationally designed taking advantage of the cut-off pore size of the vasculature, i.e., enhanced

permeation and retention (EPR) effect, allowing to be used for imaging, diagnosis and therapy,

holding great promise in the cancer field [6, 7]. The possibility of being designed on demand is

one of the major advantages of such nanotools. In this reasoning, they can be used both in the

preparation of cell-tracking tools (such as migration patterns and response to physical and

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

137

chemical factors) and the visualization of delivery and distribution of drugs in real time [8].

Recently, imaging tools such as microscopy images obtained in static 2D conditions have been

used to determine abnormal cellular distribution patterns [9], which could indicate loss of

adherence, a crucial step for cancer progression. However useful, 2D culture conditions do not

satisfyingly emulate the in vivo scenario, and new information on cancer cell behavior could be

exposed under a more relevant physiological environment [10].

Microfluidic technologies have been proposed as improved in vitro tools, once they can be

projected to display complex structures within dynamic environments [11]. A microfluidic device,

such as Vena8™ biochip, is able to realize functions that are not easily imaginable in

conventional two-dimensional (2D) biological analysis, such as highly parallel, sophisticated high-

throughput analysis and single-cell analysis, in terms of morphology, secretome and the influence

of growth factors in a well-defined manner that could be beneficial in cancer research [12, 13].

Its microfluidic channels on a micrometer and sub-micrometer scale are made to mimic human

capillaries and allow controlling the flow forces, the mixture of solutions, and the supply of

nutrients and biochemical agents in gradients. Moreover, we hypothesized that this model with

particular in vivo like features such as the presence of shear stress, flow and confinement, will

shed light on what happens when nanoparticles are injected in vivo and disclose different

interactions from those observed in traditional 2D in vitro assays. Other advantage of

microfluidics is the inherent miniaturization with a reduction in cell density and amount of

material needed for an experiment, which can increase efficiency when working with primary

cells, while testing different concentrations of extracts or drugs can be performed simultaneously

[14, 15]. In fact, the combination of microfluidics with tissue engineering to study cancer has

been recently reported with very promising results in angiogenesis and metastasis [16, 17].

These two mechanisms are essential to understand the complex development of cancer [18, 19],

by providing new data on crosstalk among several cell types.

The association of nanoparticles tracking systems with microfluidics holds great promise as

an accurate platform for the in vitro validation of new therapies, and consequently expediting the

translation of those therapies into the clinic. In fact, a major problem that has been hampering

this translation is the lack of proper in vitro models capable of emulating the in vivo environment,

including all the barriers naturally present, such as the blood flow and the interactions with the

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

138

endothelium and with cells, which are not usually taken in consideration, and the absence of

fluorescent-labeled drug delivery systems that prevents the monitoring of therapy efficiency [6,

20, 21]. Additionally, little is known about nanoparticles effect when submitted to a

physiologically-relevant environment similar to the one that cells are surrounded of, further

highlighting limitations of the conventional 2D culture conditions. In the work of Fede and

colleagues, it is possible to observe the effect of directional flow, as observed in blood vessels, in

the cellular uptake of gold nanoparticles, and their non-cytotoxicity as compared with cells

cultured under static conditions [22]. To address the aforementioned problems, we hypothesized

that the use of labeled nanoparticles together with a microfluidic platform is suitable for cancer

cell and drug delivery system tracking in real time, which better mimics the in vivo

microenvironment than traditional static cultures. This platform works as a transitional model for

the assessment of dynamic nanoparticle internalization and cancer cell tracking, which will be

useful in subsequent studies with 3D models. The labeled nanoparticles were achieved by

grafting the fluorescent label probe Fluorescein-5(6)-isothiocyanate (FITC) to

poly(amidoamine)/carboxymethilchytosan (CMCht/PAMAM) dendrimer nanoparticles. It is

noteworthy that dendrimer nanoparticles can be easily labeled with different fluorochromes,

depending on the need for combining different co-stainings. In this work, surface modified

dendrimers were used due to its remarkable physicochemical and biological properties, and have

great potential to be used in fundamental cell biology [23-25]. Moreover, since the biological

properties of poly(amidoamine) dendrimers can be tuned with respect to their terminal

functionality, it is important to investigate the biocompatibility of any new classes of derivatives of

PAMAM dendrimers proposed for biological applications.

The modified nanoparticles were characterized with several physicochemical techniques,

such as dynamic light scattering (DLS), proton nuclear magnetic resonance (H1 NMR),

transmission electron microcopy (TEM), atomic force microcopy (AFM) and differential scanning

calorimetry (DSC). Cytotoxicity was screened using three different human cancer cell lines:

U87MG (glioblastoma); HeLa (cervical cancer) and HCT-116 (colon cancer) by means of

performing a tetrazolium reduction (MTS) assay. Proliferation studies were carried out using a

DNA quantification assay. Cancer cell lines were exposed to different concentrations of

CMCht/PAMAM dendrimer nanoparticles over a period of 3 days in conventional 2D standard

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

139

conditions. After finding the highest non-cytotoxic concentration, the internalization efficiency was

assessed, both qualitatively and quantitatively (using fluorescence microscopy and flow

cytometry, respectively), in static 2D cultures and dynamic conditions using Vena8™ biochips.

V-3. MATERIALS AND METHODS

V-3.1. Synthesis of the CMCht/PAMAM dendrimer nanoparticles

CMCht with a degree of deacetylation of 80 % and degree of substitution of 47 % was

synthesized by a chemical modification route of chitin (Sigma, Germany) as described by Chen et

al.[26].

Starburst PAMAM carboxylic acid terminated dendrimers, hereafter designated as PAMAM-

CT (generation 1.5, 20 %wt. methanolic solution) with an ethylenediamine core were purchased

(Sigma-Aldrich). CMCht/PAMAM dendrimer nanoparticles were prepared in a step-wise manner

as follows: i) increase the generation of the PAMAM-CT (G1.5), ii) obtain a PAMAM methyl ester

terminated dendrimer, iii) react PAMAM and CMCht (the reaction occurs by a condensation

reaction between the methyl ester and amine groups), and iv) convert methyl ester groups, that

do not react, into carboxylic groups in the CMCht/PAMAM dendrimer, followed by precipitation.

First, the increase of the dendrimers’ generation was carried out. For that, an appropriate volume

of PAMAM-CT (G1.5) in methanol was transferred to a volumetric flask and the solvent

evaporated off under nitrogen gas, and the traces dried under vacuum in order to completely

remove the methanol. The starting compound was re-dissolved in ultra-pure water to give a final

concentration of 10 mg/mL and the pH was adjusted to 6.5 with dilute hydrochloric acid solution

(Riedel de-Haen, Germany). 1-Ethyl-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC,

Fluka) was then added to the solution at a molar ratio sufficient to modify the carboxylate residue

of the dendrimers, and the solution was kept under agitation for 30 minutes at room temperature

(RT). Ethylenediamine (EDA, Sigma, Germany) was added to the solution at a molar ratio equal to

that of EDC and left to react for at least 4 hours. After this period the excess of EDC was removed

by dialysis (cellulose tubing, benzoylated for separating compounds with a molecular weight of

≤1200, Sigma, Germany). The compound was used without purification in the next step. After

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

140

preparing the PAMAM-amine terminated compound (PAMAM-AT) an exhaustive alkylation of the

primary amines (Michael addition) was performed. An appropriate volume of PAMAM-AT was

mixed with methanol (Sigma, Germany) and methyl methacrylate). The solution was kept under

agitation in a water bath for 24 hours at 50 °C, to obtain the PAMAM-methyl ester. The CMCht in

ultrapure water was mixed with the PAMAM-methyl ester dendrimer, which was previously

dissolved in a 20/80 water/methanol (v/v) solution. The final solution was diluted by adding

methanol and kept under agitation for 72 hours. After this period, CMCht/PAMAM dendrimers

with carboxylic-terminated groups were obtained as described elsewhere [27]. CMCht/PAMAM

dendrimer nanoparticles were then precipitated after addition of an appropriate volume of a

saturated Na2CO3 (Aldrich, Germany) solution and cold acetone (Pronalab, Portugal). Precipitates

were collected by filtration and dispersed in ultrapure water for dialysis over a period of 48 hours.

CMCht/PAMAM dendrimer nanoparticles were obtained by freezing the solution at 80 °C and

freeze-drying (Telstar-Cryodos-80, Portugal) up to 7 days to completely remove the solvent. It is

worth noting that the CMCht/PAMAM dendrimer nanoparticles are water-soluble at physiological

pH.

V-3.2. Labeling of CMCht/PAMAM dendrimer nanoparticles with FITC

The conjugates of CMCht/PAMAM-FITC were obtained by covalently bond the amine group

of CMCht and the isothyocianate group of FITC (10 mg/mL FITC (Sigma, Germany) in anhydrous

dimethyl sulfoxide (DMSO, Norconcessus) creating a thiourea bond. First, a 10 mg/mL

CMCht/PAMAM dendrimer nanoparticle solution was prepared in a carbonate-bicarbonate

coupled buffer of pH 9.2. Then, a solution of FITC/DMSO was added under agitation and kept in

the dark at 4 °C for 8 hours. At this moment, the FITC-labeled CMCht/PAMAM dendrimer

nanoparticle solution was dialyzed against ultra-pure water in order to remove unlinked FITC for

24 hours and filtered (pore size <220 nm) in sterile and dark conditions. The final product was

obtained after freeze-drying.

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

141

V-3.3. Characterization of CMCht/PAMAM dendrimer nanoparticles

V-3.3.1. Transmission electron microscope (TEM)

The morphology was investigated by transmission electron microscopy, TEM (Philips CM-12,

FEI Company, The Netherlands, equipped with a MEGA VIEW-II DOCU camera and Image

Software Analyzer SIS NT DOCU). For that, the nanoparticles were stained with 2 % of

phosphotungstic acid and placed on copper grids for observation.

V-3.3.2. Atomic force microscopy (AFM)

The morphology of the nanoparticles was also investigated using atomic force microscopy

(AFM). First, the freeze-dried CMCht/PAMAM dendrimer nanoparticles were dispersed in

ultrapure water to obtain a solution with final concentration of 1 mg/mL and then one drop was

placed over a 9.9 mm mica disc (Agar Scientific, England) and blown dried with nitrogen gas for

subsequent characterization. Then, the samples were analyzed using the Tapping Mode™ with a

MultiMode AFM connected to a NanoScope III controller, both from Veeco, USA, with non-contact

silicon nanoprobes (ca 300 kHz) from Nanosensors, Switzerland. All images were plane-fitted

using the third degree-flatten procedure included in the NanoScope software version 4.43r8. The

morphometric analysis was performed using the same software.

V-3.3.3. Dynamic light scattering (DLS)

Zeta potential and particle size of the CMCht/PAMAM dendrimer nanoparticles were

measured in a particle size analyzer (Zetasizer Nano ZS, Malvern Instruments, UK). Particle size

analyses were performed by dynamic light scattering (DLS), in an aqueous solution with low

concentration of nanoparticles and using disposable sizing cuvettes. Electrophoretic

determinations of Zeta potential were investigated using the universal ‘dip’ cell pH 7.4 in

phosphate buffered saline (PBS) solution. Zeta potential was also investigated in water.

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

142

V-3.3.4. Nuclear magnetic resonance (NMR) spectroscopy

In this work, for the determination of the structure of the prepared CMCht/PAMAM

dendrimer nanoparticles, it was used H1NMR analyses. For this purpose, nanoparticles were

dissolved in deuterated water (D2O). Then, the NMR spectra were obtained with a Mercury -

400BB operating at a frequency of 399.9 MHz at 50 °C. The one-dimensional 1H spectra were

acquired using a 45° pulse, a spectral width of 6.3 kHz and an acquisition time of 2.001

seconds.

V-3.3.5. Differential scanning calorimetry (DSC)

Differential scanning calorimetry (DSC) was performed to study the thermal stability and

changes in crystallinity over a range of temperatures. A known mass of powder was placed in an

aluminum pan and a lid was crimped onto the pan. The pan was then placed in the sample cell

of a DSC module. The temperature of the DSC module was equilibrated at 35 °C and then

increased at a rate of 10 °C per minute. under a N2 gas purge until the material began to

degrade. The temperatures were obtained for each peak in the resulting curve and provided

indications of temperature stability and phase transitions.

V-3.4. In vitro studies under static conditions

V-3.4.1. Cell culture

Three cell lines were used: U87MG cells (human glioblastoma cell line), HeLa cells (human

cervical cancer cell line) and HCT-116 cells (human colon cancer cell line) were originally

obtained from the American Collection of Cell Cultures (ATCC, USA). Cells were continuously

grown in Dulbecco’s Modified Eagle Medium (DMEM, Gibco, Invitrogen) supplemented with 10 %

(v/v) fetal bovine serum and 1 % (v/v) penicillin and streptomycin under standard conditions (37

°C in a humidified atmosphere containing 5 % CO2). Subcultures of cells were performed when

confluence reached values of ~90 %. Then, cells were plated at an initial density of 2x104

cells/well in 24-well plates and exposed to three concentrations of CMCht/PAMAM dendrimer

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

143

nanoparticles (0.1, 0.5 and 1 mg/mL) for short-term exposures comprised between 1 and 3

days.

V-3.4.2. Cell viability

For the studied cell types (HeLa, U87MG and HCT-116), cell viability was assessed by the

MTS [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2(4-sulfophenyl)-2H-tetrazolium] test

(VWR, Portugal). At each time point, cell culture medium was replaced by culture medium

containing MTS in a 5:1 ratio and incubated for 3 hours. Then, 100 µL of solution from each well

were transferred to 96-well plates and the optical density was determined at 490 nm. Cultures

that were not exposed to the CMCht/PAMAM dendrimer nanoparticles were used as controls.

V-3.4.3. DNA Quantification

To assess the effect of CMCht/PAMAM dendrimer nanoparticles on cancer cells, the total

amount of DNA was measured using the fluorimetric double-stranded DNA (dsDNA) quantification

kit (Quant-iT™ PicoGreen® dsDNA Assay Kit, Molecular Probes). HeLa, U87MG and HCT-116

cancer cells were washed with sterile PBS and were lysed with 1 mL of ultrapure water in each

well. The cellular suspensions were transferred into microtubes and incubated in a water bath at

37 °C for 1 hour, then stored at -80 °C for further analysis. DNA standards were prepared with

concentrations varying between 0 and 2 g/mL. Next, 28.7 L of sample or standard, 71.3 L

of PicoGreen solution, and 100 L of Tris–HCl–EDTA buffer were mixed in each well of an

opaque 96-well plate (Alfagene, Portugal) and were incubated in the dark for 10 minutes. After,

fluorescence was measured using an excitation wavelength of 485 nm and an emission

wavelength of 528 nm. A DNA standard curve was prepared with concentrations varying between

0 and 2 g/mL and sample DNA values were read off from the standard graph.

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

144

V-3.5. In vitro studies under a microfluidic platform

V-3.5.1. Vena8™ biochips

The Vena8™ biochip from Cellix® (Irland) is constitute by 8 channels with 120 μm of

height, 800 μm of width and 2.8 cm of length. Each channel is connected to two microwells (one

in each end) where a microfluidic recirculating pump controlled by an iPod Touch (Kima™ pump)

is connected. The sample volume of each channel was 10 µL. The chip was fabricated in

optically clear acrylic with a substrate thickness of 500 μm, enabling the observation of cells

using brightfield, phase contrast imaging and fluorescence microscopy.

V-3.5.2. Cell culture

A semi-automated microfluidic platform, Vena8™ biochip with glass coverslip (Tebu-Bio,

Portugal) was used to mimic physiological flow conditions. Each biochip channel was coated with

fibronectin (Sigma, Germany) and then placed in a humidified sterile Petri dish, incubated at 37

°C for 1.5 hours. After the incubation period, 15.000 cells (HeLa, U87MG or HCT-116) were

gently added into each microchannel and kept in a sterile Petri dish in the incubator for two

hours to allow cell adhesion. To prevent drying, after the initial 30 minutes of incubation,

complete medium DMEM was added to the channels. Three hours after, cells were exposed to

0.5 mg/mL of CMCht/PAMAM dendrimer nanoparticles for short-term exposures comprised

between 1 and 3 days. Medium was added 3 times per day.

V-3.5.3. Cell viability

Cells were seeded at an initial density of 15.000 cells/channel (day 1 time point) and

10.000 (Day 3 time point) and exposed to 0.5 mg/mL of CMCht/PAMAM dendrimer

nanoparticles for short-term exposures comprised between 1 and 3 days. The effect of

CMCht/PAMAM nanoparticles on cell viability was tested with medium renewal 3 times per day.

Cells cultured in complete culture medium were used as controls.

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

145

To test cell viability using MTS method, cells from each channel were detached using Tryple

Express (Alfagene, Portugal). Cells were aspirated and collected in a 1.5 mL tube. After

centrifugation (5 minutes, 1500 rpm) cells were incubated with culture medium containing MTS

in a 5:1 ratio and analyzed as aforementioned for 2D standard cell cultures.

V-3.5.4. DNA quantification

To assess the effect of CMCht/PAMAM dendrimer nanoparticles on cancer cells in the

microfluidic chip, the total amount of DNA was measured using Quant-iT™ PicoGreen® dsDNA

Assay Kit. HeLa, U87MG and HCT-116 cancer cells were plated at a density of 15.000

cells/microchannel and incubated with CMCht/PAMAM dendrimer nanoparticles at a

concentration of 0.5 mg/mL up to 1 to 3 days. Cells cultured in complete culture medium were

used as controls. After the incubation, cancer cells from each channel were washed with sterile

PBS and detached using Tryple Express Enzyme (Alfagene, Portugal). Cells were aspirated and

collected in a 1.5 mL tube and lysed with 1 mL of ultrapure water. The cellular suspensions were

incubated in a water bath at 37 °C for 1 hour, then stored at -80 °C for further analysis as

described previously.

V-3.6. Assessment of internalization efficiency in 2D standard cultures and using a microfluidic

platform

V-3.6.1. Fluorescence microscopy

To assess the internalization and possible cell morphological changes at each time point, a

cell suspension was prepared and seeded on TCPS coverslips (Sarstedt Inc., USA) in 24-well

plates (2 x 104 cells/well) and cultured with 0.5 mg/mL FITC-labeled CMCht/PAMAM dendrimer

nanoparticles, in standard culture conditions for 1 and 3 days. Then cells were fixed with 4 %

(v/v) formalin (Sigma, Germany) and stained for F-actin filaments of the cytoskeleton and nuclei

with Texas Red-X phalloidin (Molecular Probes, Invitrogen, USA) and with 4,6-diamidino-2-

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

146

phenylindole, dilactate (DAPI blue, Molecular Probes), respectively, following supplier’s protocol.

For dynamic cultures, cells were cultured as described previously in the microfluidic platform and

were fixed with 10 µL/channel of 4 % (v/v) formalin (Sigma, Germany) for 10 minutes. Then, 10

µL/channel of DAPI and Texas Red-X phalloidin were perfused into the microchannels to stain

the nuclei and the cytoskeleton F-actin filaments, respectively. Finally, cells were observed under

the fluorescence microscope (AxioImager Z1, Zeiss Inc., Germany).

V-3.6.2. Flow cytometry analysis

To quantify the internalization of FITC-labeled CMCht/PAMAM dendrimer nanoparticles, flow

cytometry analysis was performed. In static conditions, cancer cells were cultured in the

presence of 0.5 mg/mL FITC-labeled CMCht/PAMAM dendrimer nanoparticles in a 6-well plate

(2x105 cells/well). Cells cultured in complete culture medium were used as controls. Cells were

released from substratum and a cell strainer was used to avoid cell clusters. Afterwards, 0.5 mL

of DMEM complete medium was added to each well and samples transferred to cytometry tubes.

In dynamic conditions, cancer cells were cultured in the presence of 0.5 mg/mL FITC-

labeled CMCht/PAMAM dendrimer nanoparticles in the microfluidic chips. Cells cultured in

complete culture medium were used as controls. FACS analysis was carried out to perform

quantitative analysis on the internalization efficiency as well as cell viability, as follows: each

channel was washed with PBS. Then, PBS was aspirated and cells from each channel were

detached using 10 L of Tryple Express and collected in an FACS tube. After centrifugation at

1500 rpm for 5 minutes, cells were re-suspended in 800 L of a 2 % (v/v) fetal bovine serum in

PBS solution. Afterwards, 5 L of 7-AAD (Taper group, Portugal) was added to each sample for

determining the number of dead cells. After this step, cells were loaded in FACSCalibur flow

cytometer (BD Biosciences Immunocytometry Systems, USA). Calibrate beads three-color kit (BD

CaliBRITETM beads, USA) was used to adjust the equipment instrument settings before samples

are run on the flow cytometer. Finally, data was treated using the FLOWING SOFTWARE 2.

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

147

V-4. RESULTS

V-4.1. Characterization of the CMCht/PAMAM dendrimer nanoparticles

In order to gain deeper knowledge on the CMCht/PAMAM dendrimer nanoparticles

physicochemical properties, a TEM and AFM study were performed. Their architecture

determined in dry conditions is depicted in Figures IV-1A and I-1B. From Figure IV-1A, it is clearly

visible the monodisperse character of these particles, as well as the polymer coating around

PAMAM’s core and a clear dendron-like morphology, which was further confirmed by AFM

analysis as shown in Figure IV-1B. In fact, it is possible to observe the 3D AFM image of the

CMCht/PAMAM dendrimer nanoparticles, which noticeably shows the consistency of these

molecules, i.e. nanosphere-like shape. By assessing the particle distribution using DLS, it

revealed that the CMCht/PAMAM dendrimer nanoparticles have an average size of 50 nm

(Figure IV-1C). After the synthesis of the CMCht/PAMAM dendrimer nanoparticles, three types of

particles were clearly detected, with sizes approximately 6 ± 1 nm, 54 ± 7 nm and 674 ± 462

nm. The first graph corresponds to the modified dendrimer nanoparticles with

carboxymethylchitosan where two peaks are present, corresponding to single and aggregated

nanoparticles, respectively. The second graph corresponds to the unmodified PAMAM dendrimer

of generation 1.5, also with two visible peaks. These correspond to single and aggregated

nanoparticles. Zeta potential data shows that nanoparticles dispersed in water at neutral pH

present negative charge of -34.3 ± 3 (mV).

Figure IV-1D shows the H1NMR spectrum of the synthesized CMCht/PAMAM dendrimer

nanoparticles and its comparison with unmodified PAMAM dendrimer and biocompatible polymer

carboxymethylchitosan, revealing a singlet at 2.49 ppm, and multiplets from 3.1 to 3.73 ppm

and 4.07 ppm associated to the H2 protons, the ring methine protons (H3, H4, H5 and H6), and

protons of –CH2COO- groups of the CMCht. Also, peaks at 2.49 and 3.46 ppm appear

overlapped as a result of the resonances of the CH2COO and CH2 protons of both CMCht and

PAMAM. The presence of the peak at ca. 3 ppm and the absence of a peak at 4.3 ppm can

indicate that the substitutions occurred mainly on the C2 amino group of CMCht. Finally, DSC

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

148

thermogram analysis to assess thermal stability of the modified dendrimer shown in Figure IV-1E

reveals a Glass transition temperature (Tg) of -13.7 °C and a thermal decomposition at 267 °C.

Figure V-1 - Characterization of water soluble nanoparticles. A) Transmission electron microscopy (TEM) image of the

CMCht/PAMAM. B) Atomic force microscopy (AFM) images of CMCht/PAMAM on a quartz mica. C)

Particle size distribution (nm). D) H1NMR spectra of CMCht/PAMAM, PAMAM G1.5 and CMCht. E)

Differential Scanning Calorimetry (DSC) spectrum of CMCht/PAMAM.

V-4.2. Cell viability/proliferation in static conditions

Cell viability and proliferation were determined by means of performing MTS and DNA

quantification assays (Figure IV-2). The three chosen cancer cell lines were exposed to a range of

dendrimer concentrations over a period of 3 days. Our results indicated showed no deleterious

effects on cell viability in static culture conditions, when comparing the tested and control

samples (p<0.0001). Therefore, the tested conditions were not affected significantly by the

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

149

presence of CMCht/PAMAM dendrimer nanoparticles. It is clear that all cancer cell lines

proliferate in a significant manner from day 1 to day 3 in the absence of nanoparticles (controls).

As depicted in Figure IV-2, the concentration with less variability, more consistent results (low

standard deviations), and more alike to the control group (no statistically significant differences)

was 0.5 mg/mL. This preliminary tests allowed us to screen three types of relevant cancer cell

lines with a wide range of concentrations and choose the most adequate one to proceed to

comparative studies between static and dynamic conditions of culture, in order to validate a

model for real time tracking. So, from this point on, all the tests were performed with a

dendrimer nanoparticle concentration of 0.5 mg/mL.

Figure V-2 - Normalization of MTS results by DNA results of HCT-116, HeLa and U87MG human cancer cell lines in

static conditions in the presence of CMCht/PAMAM dendrimer nanoparticles. Assays were conducted

to assess the cytotoxicity effect of nanoparticles on cell viability at concentrations of 0.1, 0.5 and 1

mg/mL at time points of 1 and 3 days (* indicates significant differences when comparing the

different concentrations to control (0 mg/mL) at each time point. Ф indicates significant differences

when comparing controls from day 3 to day 1 to determine cell proliferation).

V-4.3. Cell viability/proliferation in dynamic conditions

Figure IV-3 shows the results of the two tested groups (static and dynamic) at day 1 and day

3, in the absence or presence of nanoparticles (0.5 mg/mL), for easy comparison. The results

shown represent the normalization between the two tests. Regarding the dynamic condition, HCT-

116 cells showed greater sensitivity in the presence of nanoparticles, both in day 1 and day 3,

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

150

which is translated into a decrease of viability. U87MG cells, by its turn, showed this significant

difference between control and nanoparticles groups only at day 3. However, for HeLa cells, no

significant differences were observed in dynamic conditions regarding the effect of nanoparticles.

Figure V-3 - Cytotoxicity of CMCht/PAMAM over HCT-116, HeLa and U87MG human cancer cell lines in static and

dynamic conditions. Assays were conducted to assess cytotoxicity/viability over the cells at a

concentration of 0.5 mg/mL at different time points (days 1 and 3) and it in static and dynamic

conditions. (* indicates significant differences when comparing 0.5 mg/mL to control (0 mg/mL) at

each time point. Ф indicates significant differences when comparing controls from day 3 to day 1 to

determine cell proliferation).

V-4.4. Assessment of nanoparticles cellular uptake in 2D standard cultures

V-4.4.1. Fluorescence microscopy and flow cytometry analysis

Figure IV-4 shows the fluorescence microscopy images obtained after culturing the three

cancer cell lines with 0.5 mg/mL of FITC-CMCht/PAMAM for 1 and 3 days in static conditions.

As can be observed, all cancer cells were able to internalize the fluorescent labeled

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

151

nanoparticles. Nanoparticles are dispersed in the cytoplasm but especially around the nucleus

(stained with DAPI). Also, it is clear that the internalization rate tended to increase with time. The

microscopic observation of cells showed no relevant cell morphological changes, which

corroborate the lack of cytotoxicity.

Figure V-4 - Fluorescent microscope images of FITC-labeled CMCht/PAMAM dendrimer nanoparticles (green)

internalization in cancer cells labelled with DAPI (nuclei) and Phalloidin (F-actin, cytoskeleton) after 1

and 3 days in direct contact with dendrimer nanoparticles in static conditions.

Quantitative analysis of nanoparticles uptake at the cellular level was performed using flow

cytometry analysis. Table IV-1 shows the percentage of internalization of FITC-labeled

CMCht/PAMAM dendrimer nanoparticles by each cell line under static conditions, as well as the

percentage of live cells obtained by staining with 7-AAD, determined by flow cytometry. Flow

cytometry results of cancer cell lines cultured in the presence of FITC-labeled CMCht/PAMAM

dendrimer nanoparticles showed that HeLa cancer cells continued to internalize the

nanoparticles along time, starting with an internalization percentage of 87 % by day one, and

increasing to 99 % by day 3 (higher fluorescent signals along time). By its turn, U87MG and HCT-

116 internalization rates were constant and reached the maximum peak, as almost 100 % of

cells showed internalized nanoparticles only after 24 hours. Regarding cell viability in static

conditions, it is possible to observe that cells remain viable in the presence of dendrimer

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

152

nanoparticles (% of live cells on day 1 is very similar to day 3) and no relevant cytotoxic effects

are noted when compared to controls (See Table IV-1).

Table V-1 - FACS data of the percentage of internalization of FITC-labeled CMCht/PAMAM dendrimer nanoparticles

(NP) and the percentage of live cells (stained with 7AAD), from day 1 to day 3 in static conditions.

V-4.5. Assessment of internalization efficiency using a microfluidic platform

V-4.5.1. Fluorescence microscopy and flow cytometry analysis

Figure IV-5 represents a panel of images with results obtained after 1 and 3 days of culture

in the presence of FITC-CMCht/PAMAM at a concentration of 0.5 mg/mL, in dynamic culturing

conditions. As can be observed, all cancer cells were able to internalize the fluorescent-labeled

nanoparticles when cultured in the microfluidic system.

Culture conditions Internalization (%) Live cells (%)

Static Day 1 Day 3 Day 1 Day 3

HeLa (NP) 87 99 96 97

HeLa (Control) 0 0 99 99

U87MG (NP) 98 98 97 95

U87MG (Control) 0 0 99 99

HCT-116 (NP) 100 99 97 97

HCT-116 (control) 0 0 99 100

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

153

Figure V-5 - Fluorescent microscope images of FITC-labeled CMCht/PAMAM dendrimer nanoparticles (green)

internalization in cancer cells labeled with DAPI (nuclei) and Phalloidin (F-actin, cytoskeleton) after 24

hours and 72 hours in direct contact in dynamic conditions.

Flow cytometry studies on cells cultured in dynamic conditions reveal similar results to static

conditions when it comes to nanoparticles’ internalization performance. Cells are able to

internalize nanoparticles almost at the maximum rate by 24 hours after the beginning of the

experiment (Table IV-2). Concerning cell viability, it is possible to observe differences between the

absence and the presence of nanoparticles. In fact, all cancer cells show decreased viability in

the presence of nanoparticles. Special emphasis in HCT-116 with a percentage of live cells of 90

% when compared to the respective control (99 %), followed by U87MG with 95% against 99 % of

the control, and finally HeLa cells with a percentage of 97 % in the presence of nanoparticles and

100 % in the control. It is worth noticing that the viability in relation to control was kept when

analyzing day 3 (Table IV-2).

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

154

Table V-2 - FACS data of the percentage of internalization of FITC-labeled CMCht/PAMAM dendrimer nanoparticles

and the percentage of live cells (stained with 7AAD), from day 1 to day 3 in dynamic conditions.

V-5. DISCUSSION

Although the cancer research field has been evolving along the years, there is still missing a

proper tool to track cells and therapy efficiency in real time, allowing the interpretation of their

interaction and expediting its translation into the clinics. Cell movement, location and

quantification studies require specialized probes that are non-cytotoxic and are available in a

range of fluorescent colors to match instrument lasers and filters, and to accommodate co-

staining with antibodies or other cell analysis probes.

In light of this, the main goal of this work was aimed at developing a new platform

composed of fluorescent-labeled nanoparticles associated with a microfluidic device, which may

be useful for the validation of new theranostic approaches. To accomplish this, we investigated

the interactions between a new class of labeled nanoparticles, developed previously by our group

[28] with several cancer cell lines, both in standard 2D static culture systems and in a dynamic

microfluidic chip. The in vitro studies demonstrated that cells remain viable in both culturing

conditions. However, cell cultures in the dynamic systems displayed higher sensitivity to

nanoparticles, disclosing the real importance of a more physiologically relevant model for the

development of new cancer therapies, otherwise, no differences in cells response should be

found. Additionally, as the developed nanoparticles show internalization rates near to 100 % (all

the cells in culture have internalized fluorescent labeled nanoparticles) after 24 hours in culture,

they enable real time monitoring of drug delivery systems efficiency accelerating its translation

into clinics.

Culture conditions Internalization (%) Live cells (%)

Dynamic Day 1 Day 3 Day 1 Day 3

HeLa (NP) 95 100 97 97

HeLa (Control) 0 0 100 100

U87MG (NP) 99 99 95 92

U87MG (Control) 0 0 99 98

HCT-116 (NP) 100 99 90 90

HCT (Control) 0 0 99 99

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

155

Recently, our group has proposed a strategy to chemically modify PAMAM dendrimer

nanoparticles with carboxymethylchitosan in order to achieve water solubility and low cytotoxicity,

making them ideal for this purpose [28]. The first steps were to synthetize and modify the

nanoparticles and to physico-chemically characterize them, prior to its use in biological studies.

The chosen methods are simple and reliable tools to determine chemical composition,

morphological features and surface chemistry, such as TEM, AFM, DLS, H1NMR and DSC

methods. In this study, the TEM analysis (Figure IV-1A) clearly revealed a dendron-like

morphology with a central core and the added carboxymethilchitosan polymer coating. This data

was further confirmed by AFM analysis (Figure IV-1B) that visibly showed the consistency of these

molecules features, i.e., nanosphere-like shape, which is in agreement with previously described

results [28, 29].

CMCht/PAMAM dendrimer nanoparticles were also analyzed with DLS. Particle size is one

of the most important parameters in drug carriers for intracellular delivery, as well as the surface

charge, which was also measured. Three sizes of particles were detected, approximately 6 ± 1

nm, 54 ± 7 and 674 ± 462 (Figure IV-1C). In the first graph, where is depicted the analysis of

CMCht/PAMAM dendrimer, the first peak corresponds to the modified dendrimer and the second

to the aggregates. Concerning the second graph, the first peak matches the size of the isolated

unmodified PAMAM G1.5 dendrimer nanoparticles and the second to the size of aggregated

nanoparticles. The last group, the aggregated nanoparticles, can be explained by interactions

between non-covalent free carboxyl groups of CMCht and/or unmodified PAMAM-Carboxyl

terminal dendrimer and amino groups of other CMCht particles, which may promote the

formation of nanoparticle aggregates [30]. This is a property involving not only the surface

properties of the particles but also their environment, e.g. pH, ionic strength, and even the type

of ions in the suspension. By its turn, it is expected that CMCht has a good pH and ion sensitivity

in aqueous solution due to abundant –COOH and –NH2 groups. These results corroborate the

theory that the negatively charged carboxymethyl groups in the CMCht/PAMAM dendrimer

nanoparticles are mainly distributed at the surface of the nanoparticles working as a water-

soluble coating, as proved by the DLS data.

Regarding the H1NMR analysis, the singlet at 2.49 ppm and multiplets from 3.1 to 3.73

ppm and 4.07 ppm are associated to the H2 protons, the ring methine protons (H3, H4, H5 and

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

156

H6), and protons of –CH2COO- groups of the CMCht, respectively [30, 31]. Also, the presence of

peaks from 1.98 to 3.46 ppm is noteworthy because they point the presence of PAMAM

dendrimer. These have been attributed to CH2COO, CH2CO and CH2CN groups. The presence of

the peak at ca. 3 ppm and the absence of a peak at 4.3 ppm can indicate that the substitutions

occurred mainly on the C2 amino group of CMCht, confirming the modification of the PAMAM

dendrimer. SC was used to evaluate the thermal stability of DNA complexed to PAMAM

dendrimers SC was used to evaluate the thermal stability of DNA complexed to PAMAM

dendrimers. The DSC studies were performed to assess the thermal stability of PAMAM

dendrimers modified with carboxymethilchitosan. The thermal degradation was observed at 267

°C indicating that the developed dendrimers are high thermally stable at body temperature. This

high value could be due to the presence of hydrogen bonds in the structure of PAMAM, as

observed previously [32]. Nevertheless it is lower than the observed in previous studies made

with other generations of PAMAM dendrimers [33]. This variation can be explained by the

contribution of carboxymethilchitosan polymer. In fact, it was previously shown that the longer

the grafted polymers are, the lower the values of thermal degradation obtained, and consequently

less stable the dendrimer is [34]. Moreover, a thermal glass transition is clear at -13.7 °C. To the

best of our knowledge, this type of characterization has never been done in this specific type of

modified dendrimers and therefore could serve as a reference for future works.

Regarding the biological assays, similarly to what has been done by Oliveira et al. for

neurons, glial cells [29], fibroblasts (L929) and rat bone marrow stromal cells (RBMSCs) [28],

the cytotoxic effect of different concentrations of CMCht/PAMAM dendrimer nanoparticles on

glioblastoma cancer cells, colon cancer cells and cervical cancer cells, along the time, was the

first parameter to be assessed, as a bridge to dynamic experiments. This allowed us to select the

higher non-cytotoxic concentration to be used in subsequent assays, which was 0.5 mg/mL.

After that, standard 2D culture systems in cell culture TCPS well-plate (static conditions) and

cultures inside a microchannel in a microfluidic device (dynamic conditions) were comparatively

investigated. The last culture system works as a bridge towards more elaborated 3D models, as it

mimics the natural environment that cells and nanoparticles are exposed to, once in systemic

circulation in vivo [35], regarding confinement, flow and shear stress. Figure IV-2 represents the

results of the screening under static conditions, which indicates highly metabolically active cells.

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

157

HCT-116, HeLa and U87MG proliferated well and were viable either in the controls (absence of

nanoparticles) or in the presence of nanoparticles at any given concentration. Statistically, it is

possible to observe that increasing concentrations of CMCht/PAMAM (0.1, 0.5 and 1 mg/mL)

did not affect the viability or proliferation of cancer cell lines. This can be explained by the natural

aggressiveness and resistance of cancer cell lines [36] when compared to other cell types as

showed in the work of Oliveira et al [28, 30]. For so, we decided to use the intermediate

concentration: 0.5 mg/mL as it shows no statistical differences when compared to the controls,

and because it could potentially increase the number of nanoparticles inside the cells when

compared to the lowest concentration (0.1 mg/mL).

When static conditions were compared to dynamic conditions (Figure IV- 4), in which cells

were exposed to 0.5 mg/mL for the same periods of time, HCT-116 cells showed a clear

sensitivity to nanoparticles, especially at day 3, as well as U87MG cells. None of these

phenomena were observed in static standard conditions, highlighting the importance of using a

system that mimics features of the in vivo environment. Therefore, the main differences between

the two culture conditions rely on the fact that nanoparticles show no effect on any of the chosen

cancer cell lines at any time point in static conditions. Nevertheless, when the same cells were

cultured in a confined and dynamic environment they showed to be more sensitive. Moreover, we

observed that their growth is much greater in static conditions (statistic differences from day 1 to

day 3) which can be due to the fact that they proliferate freely in traditional 2D culture conditions,

on contrary to what was observed in dynamic conditions due to the limited space available within

the microchannels of the microfluidic device. This is why it is also worth noticing the short time

exposures (3 days) used in this work instead of the traditional 14 days’ exposure. In fact,

microfluidic devices, such as Vena8™ chips, are made to speed up and scale down experiments

[37].

Qualitative images obtained with fluorescence microscope revealed that either in static and

dynamic conditions, HCT-116, U87MG and HeLa cells were able to broadly internalize

fluorescence labeled nanoparticles at a very quick rate after 1 day in culture. The normal

morphology of the cells (marked by DAPI and Phalloidin) suggests healthy proliferation and

survival under both conditions (Figure IV-4 and Figure IV- 5). It is not clear whether the particles

are also located inside the nucleus just by observing the microscope images. Further

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

158

experiments are required to clarify the fate of these nanoparticles. The same effect has been

observed earlier by Oliveira et al. [28] in L929 and Saos-2 cell lines.

These findings were latter corroborated by flow cytometry, which was used to quantitatively

study the levels of FITC-labeled CMCht/PAMAM dendrimer nanoparticles internalized by cells.

Quantitative analysis of nanoparticles uptake at the cellular level is critical for a realistic

evaluation of their effects and to compare possible different behaviors among different cell

culture times. FACS analysis was performed to circumvent this limitation since each cell line

interacts differently with nanoparticles [38].

Results from Table IV-1 revealed increasing levels of fluorescence for HeLa cells from day 1

(87 %) to day 3 (99 %), in static conditions. By its turn, U87MG and HCT-116 cells’ internalization

rates were constant and reached the maximum peak, as almost 100 % of cells showed

internalized nanoparticles only after 1 day. A similar trend was observed in dynamic conditions

(Table IV-2). Although HeLa cells showed slower internalization rates, as compared to U87MG

and HCT-116 cells, a higher rate of internalization was observed at day 1 when comparing to

static conditions (87 % in static vs 95 % in dynamic), which could influence for example the

optimization of drug concentration used, showing once again the relevance of using a dynamic

system for validation of new therapies. The fact that HeLa cells take more time to internalize

nanoparticles in both culture conditions could be related and possibly explain their higher

sensitivity. Nevertheless, it is important to point out that almost 100 % of cells were labeled after

1 day, indicating that the developed nanoparticles present great potential as cell tracking system.

Additionally, we used flow cytometry to measure the viability levels of cells in both conditions

(static vs dynamic) using 7-AAD. 7-AAD appears to be generally excluded from live cells, creating

complexes only with free DNA from dead cells, making it an excellent quantitative indicator of cell

viability, as it gives fluorescent signal for dead cells only. By quantifying the percentage of live

cells, we observed that there were no significant differences between the controls and the

presence of 0.5 mg/mL of nanoparticles in static studies (Table IV-1). On the other hand, when

seeded and cultured in the microfluidic chips with dynamic flow, cells showed higher sensitivity,

displaying higher percentages of dead cells, as depicted in Table IV-2. It is worth noticing that the

results obtained by flow cytometry analysis corroborate the results of MTS/DNA tests.

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

159

In brief, these findings suggest that standard 2D assays underestimated the effect of

CMCht/PAMAM dendrimer nanoparticles, while microfluidic chips, which represent a dynamic

setting, unraveled the real cytotoxicity and internalization rates of the labeled nanoparticles [39,

40]. This could only be possible due to the combination of dendrimer nanoparticles with

fluorescent dye, which allowed real-time monitoring of cells in this new system, envisaging the

use for new therapies validation in real time. Despite the interesting data, we envision performing

deeper studies to investigate the effect of loading an anti-cancer drug into the NP’s on cancer cell

lines in the proposed microfluidic chip.

V-6. CONCLUSIONS AND FUTURE PERSPECTIVES

This study provided proof-of-concept on the use of a platform composed of microfluidic chip

together with fluorescence labeled dendrimer nanoparticles for the validation of new

chemotherapeutic agents. In fact, the results show different responses to the presence of 0.5

mg/mL dendrimer nanoparticles when comparing static to dynamic conditions, with a tendency

towards higher sensitivity when subjected to confinement, flow and shear stress (dynamic

conditions). Moreover, the observed high internalization rates of the nanoparticles can be

beneficial, making them excellent intercellular carrier of anti-cancer drugs.

Recent evidence indicates that 3D and flow models more closely resemble in vivo function.

For example, plenteous reports have shown improved functionality with 3D cultures and flow

cultures for xenobiotic metabolism competence more closely mimicking in vivo levels. Therefore,

as hypothesized by us, the proposed system has some important features representative of the in

vivo microenvironment, such as confinement, flow and shear stress. This platform, together with

the efficient ability of cell tracking and drug delivery, represents a new step towards dynamic

studies in cancer cells in 3D, namely migration and drug testing. Hence, this work further shapes

the challenges associated with the use of nanoparticles to help deliver drugs more specifically

and effectively to cancer cells as well as to help detect cancer (theranostics). Future

nanoparticles and microfluidic models can boost drug validation and the understanding of

tumorigenesis processes, essential to move the cancer research field forward in terms of

prevention, detection and treatment.

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

160

Thus, the microfluidics can enable the development of diagnostics platform and

personalized therapies, as it opens the possibility to a valuable system to test and validate new

chemotherapeutic agents in a more realistic manner, expediting its translation into clinic.

V-7. ACKNOWLEDGMENTS

F.R. Maia acknowledges ERC-2012-ADG 20120216-321266 (ComplexiTE) for her Postdoc

scholarship. J. M. Oliveira thanks Portuguese Foundation for Science and Technology (FCT) for

his distinction attributed under the FCT Investigator program (IF/00423/2012). Bruno Costa also

thanks Portuguese Foundation for Science and Technology (PTDC/SAU-GMG/113795/2009 and

IF/00601/2012 to B.M.C.), Fundação Calouste Gulbenkian (B.M.C.), and Liga Portuguesa

Contra o Cancro (B.M.C.). Mariana R. Carvalho also thanks the funding through the LA ICVS/3Bs

project (UID/Multi/50026/2013).

V-8. REFERENCES

1. Siegel, R., et al., Cancer statistics, 2014. CA Cancer J Clin, 2014. 64(1): p. 9-29. 2. Ferlay, J., et al., Cancer incidence and mortality patterns in Europe: estimates for 40 countries in 2012. Eur J Cancer, 2013. 49(6): p. 1374-403. 3. Chaffer, C.L. and R.A. Weinberg, A perspective on cancer cell metastasis. Science, 2011. 331(6024): p. 1559-64. 4. Headley, M.B., et al., Visualization of immediate immune responses to pioneer metastatic cells in the lung. Nature, 2016. 531(7595): p. 513-7. 5. Haley, B. and E. Frenkel, Nanoparticles for drug delivery in cancer treatment. Urol Oncol, 2008. 26(1): p. 57-64. 6. Valencia, P.M., et al., Microfluidic technologies for accelerating the clinical translation of nanoparticles. Nat Nanotechnol, 2012. 7(10): p. 623-9. 7. Popovtzer, R., et al., Targeted gold nanoparticles enable molecular CT imaging of cancer. Nano Lett, 2008. 8(12): p. 4593-6. 8. Carvalho, M.R., et al., Evaluating Biomaterial- and Microfluidic-Based 3D Tumor Models. Trends in Biotechnology, 2015. 33(11): p. 667-678. 9. Mestre, T., et al., Quantification of topological features in cell meshes to explore E-cadherin dysfunction. Sci Rep, 2016. 6: p. 25101. 10. Sittampalam, S., et al., Three-Dimensional Cell Culture Assays: Are They More Predictive of In Vivo Efficacy than 2D Monolayer Cell-Based Assays? Assay Drug Dev Technol, 2015. 13(5): p. 254-61.

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

161

11. Bhatia, S.N. and D.E. Ingber, Microfluidic organs-on-chips. Nat Biotechnol, 2014. 32(8): p. 760-72. 12. Zhao, L., et al., Lab-on-a-Chip for anticancer drug screening using quantum dots probe based apoptosis assay. J Biomed Nanotechnol, 2013. 9(3): p. 348-56. 13. Zhou, H., L. Zhao, and X. Zhang, In-channel printing-device opening assay for micropatterning multiple cells and gene analysis. Anal Chem, 2015. 87(4): p. 2048-53. 14. Barata, D., C. van Blitterswijk, and P. Habibovic, High-throughput screening approaches and combinatorial development of biomaterials using microfluidics. Acta Biomater, 2015. 15. Kuo, C.T., et al., Modeling of cancer metastasis and drug resistance via biomimetic nano-cilia and microfluidics. Biomaterials, 2014. 35(5): p. 1562-71. 16. Bischel, L.L., et al., Tubeless microfluidic angiogenesis assay with three-dimensional endothelial-lined microvessels. Biomaterials, 2013. 34(5): p. 1471-7. 17. Bray, L.J., et al., Multi-parametric hydrogels support 3D in vitro bioengineered microenvironment models of tumour angiogenesis. Biomaterials, 2015. 53(0): p. 609-620. 18. Bersini, S., et al., A microfluidic 3D in vitro model for specificity of breast cancer metastasis to bone. Biomaterials, 2014. 35(8): p. 2454-2461. 19. Dong, Y., et al., Microfluidics and circulating tumor cells. J Mol Diagn, 2013. 15(2): p. 149-57. 20. Chen, L., et al., Real-time monitoring of a controlled drug delivery system in vivo: construction of a near infrared fluorescence monomer conjugated with pH-responsive polymeric micelles. Journal of Materials Chemistry B, 2016. 4(19): p. 3377-3386. 21. Kwak, B., et al., Simulation of complex transport of nanoparticles around a tumor using tumor-microenvironment-on-chip. Journal of Controlled Release, 2014. 194: p. 157-167. 22. Fede, C., et al., Evaluation of gold nanoparticles toxicity towards human endothelial cells under static and flow conditions. Microvasc Res, 2015. 97: p. 147-55. 23. Kesharwani, P., et al., PAMAM dendrimers as promising nanocarriers for RNAi therapeutics. Materials Today, 2015. 18(10): p. 565-572. 24. Wolinsky, J.B. and M.W. Grinstaff, Therapeutic and diagnostic applications of dendrimers for cancer treatment. Adv Drug Deliv Rev, 2008. 60(9): p. 1037-55. 25. Pojo, M., et al., In vitro evaluation of the cytotoxicity and cellular uptake of CMCht/PAMAM dendrimer nanoparticles by glioblastoma cell models. Journal of Nanoparticle Research, 2013. 15(5): p. 1-9. 26. Chen, X.-G. and H.-J. Park, Chemical characteristics of O-carboxymethyl chitosans related to the preparation conditions. Carbohydrate Polymers, 2003. 53(4): p. 355-359. 27. Sashiwa, H., Y. Shigemasa, and R. Roy, Chemical modification of chitosan. Part 9: Reaction of N-carboxyethylchitosan methyl ester with diamines of acetal ending PAMAM dendrimers. Carbohydrate Polymers, 2002. 47(2): p. 201-208. 28. Oliveira, J.M., et al., Surface Engineered Carboxymethylchitosan/Poly(amidoamine) Dendrimer Nanoparticles for Intracellular Targeting. Advanced Functional Materials, 2008. 18(12): p. 1840-1853. 29. Salgado, A.J., et al., Carboxymethylchitosan/poly(amidoamine) dendrimer nanoparticles in central nervous systems-regenerative medicine: effects on neuron/glial cell viability and internalization efficiency. Macromol Biosci, 2010. 10(10): p. 1130-40.

Chapter V – A Semiautomated Microfluidic Platform for Real Time Investigation of Nanoparticles’ Cellular Uptake and Cancer Cell’s Tracking

162

30. Oliveira, J.M., et al., The osteogenic differentiation of rat bone marrow stromal cells cultured with dexamethasone-loaded carboxymethylchitosan/poly(amidoamine) dendrimer nanoparticles. Biomaterials, 2009. 30(5): p. 804-13. 31. Li, Y., et al., In situ crosslinkable hydrogels formed from modified starch and O-carboxymethyl chitosan. RSC Advances, 2015. 5(38): p. 30303-30309. 32. Martin-Rapun, R., et al., Ionic thermotropic liquid crystal dendrimers. J Am Chem Soc, 2005. 127(20): p. 7397-403. 33. Uppuluri, S., et al., The Properties of Dendritic Polymers I: Generation 5 Poly (amidoamine) Dendrimers. 1998, DTIC Document. 34. Hui, H., F. Xiao-dong, and C. Zhong-lin, Thermo- and pH-sensitive dendrimer derivatives with a shell of poly(N,N-dimethylaminoethyl methacrylate) and study of their controlled drug release behavior. Polymer, 2005. 46(22): p. 9514-9522. 35. Sato, M., et al., Microcirculation-on-a-Chip: A Microfluidic Platform for Assaying Blood- and Lymphatic-Vessel Permeability. PLoS One, 2015. 10(9): p. e0137301. 36. Kuete, V., et al., Cytotoxicity of a naturally occurring furoquinoline alkaloid and four acridone alkaloids towards multi-factorial drug-resistant cancer cells. Phytomedicine, 2015. 22(10): p. 946-951. 37. Teh, S.Y., et al., Droplet microfluidics. Lab Chip, 2008. 8(2): p. 198-220. 38. Mahmood, M., et al., Cytotoxicity and biological effects of functional nanomaterials delivered to various cell lines. J Appl Toxicol, 2010. 30(1): p. 74-83. 39. Carvalho, M.R., et al., Evaluating Biomaterial- and Microfluidic-Based 3D Tumor Models. Trends Biotechnol, 2015. 33(11): p. 667-78. 40. Horning, J.L., et al., 3-D tumor model for in vitro evaluation of anticancer drugs. Mol Pharm, 2008. 5(5): p. 849-62.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

163

Chapter VI

Tuning Enzymatically-

crosslinked Silk fibroin

Hydrogel Properties for the

Development of a Colorectal

cancer Extravasation 3D Model

on a Chip

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

164

Chapter VI

Chapter VI - Tuning Enzymatically-crosslinked Silk fibroin

Hydrogel Properties for the Development of a Colorectal

cancer Extravasation 3D Model on a Chip1

ABSTRACT

Microfluidic devices are now the most promising tool to mimic in vivo like scenarios such as

tumorigenesis and metastasis due to its ability to more closely mimic cell’s natural

microenvironment (such as three dimensional (3D) environment and continuous perfusion of

nutrients). In this study, the ability of 2 % and 3 % enzymatically-crosslinked silk fibroin (SF)

hydrogels with different mechanical properties were tested in terms of colorectal cancer cell

migration, under different microenvironments in a 3D dynamic model. Matrigel was used as

control. Moreover, we present a comprehensive comparison between the traditional Boyden

chamber assay and our 3D dynamic microfluidic model in terms of colorectal cancer cell

migration. Our results show profound differences between the two used biomaterials and the two

migration models, which were explored in terms of mechanical properties of the hydrogels as

well as the intrinsic characteristics of the models. Moreover, we validated the developed 3D

dynamic model by demonstrating that hVCAM-1 plays a major role in the extravasation process,

influencing extravasation rate and traveled distance. Furthermore, the developed model enables

precise visualization of cancer cell migration within a 3D matrix in response to

microenvironmental cues, shedding light on the importance of biophysical properties in cell

behavior.

Keywords: Colorectal cancer; Cell migration; Mechanical properties; Silk.

1This chapter is based on the following publication

M. R. Carvalho, F. R. Maia, S. Vieira, R. L. Reis, and J. M. Oliveira. “Tuning Enzymatically-crosslinked Silk fibroin Hydrogel

Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip”. Global Challenges, 2018, 2: 1870164.

doi: 10.1002/gch2.201700100.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

165

VI-1. GRAPHICAL ABSTRACT

VI-2. INTRODUCTION

Colorectal cancer is a major cause of morbidity and mortality worldwide, and accounts for

over 9 % of all cancer incidence. It is the third most common cancer worldwide and affects men

and women equally [1].

In order to win the battle against cancer, further advances are in great need to unveil

identification of cancer-causing agents in in vitro and in vivo animal models, as well as for the

development of personalized therapies, drug screening, and to provide insightful knowledge on

the mechanisms of tumor growth and metastasis [2]. Although in vivo animal models comprise

the complexity of the metastatic cascade in a living system, visualization of the distinct events is

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

166

nearly impossible. In fact, recent studies suggest that the correspondence between animal

models and successful clinical trials does not reach 10 % [3]. Furthermore, animal models do not

allow control of cell-cell and cell-extracellular matrix (ECM) interactions, making it difficult to really

understand the role of each stromal component in the tumorigenesis process. On the other hand,

two dimensional (2D) in vitro models have reduced physiological relevance, capturing only limited

aspects of the tumor microenvironment [4]. Therefore, 3D models, comprising the integration of

tissue engineering (TE) strategies with microfluidic technologies have sparked a breakthrough

into the design of in vitro microfluidic culture models. These better adapt to morphological

changes in tissue structure and function over time, providing a level of precision control that

could not be achieved before [5]. Microfluidics can provide useful model systems to investigate

complex phenomena under combination of multiple controllable biochemical and biophysical

microenvironments, coupled with high resolution real time imaging [6, 7].

This type of strategy does provide a powerful tool to develop more in vivo - like settings that

can be applied to study cancer. Special focus on model dimensionality and microenvironment

complexity is given, as a significant amount of experimental evidence has shown that both

mechanical and chemical stimuli from the cell microenvironment play a key role in several types

of cell behavior, namely in terms of migration [8]. Migration is frequently used as broad term in

biology and applies to any directed cell movement, even in a 2D setting [9]. The ability to migrate

allows cells to change their position within tissues or between different organs. However, in

pathology, invasion is defined as the crossing of (tissue) 3D barriers, into the underlying

interstitial tissues by malignant tumor cells [9].

So far, several synthetic and natural hydrogels have been used as ECM - like materials in the

in vitro cell culture studies, offering characteristics such as biocompatibility and bioactivity, and

have undoubtedly been proven to influence cells’ behavior and fate [10-12]. However, the full

potential of these recent in vitro models in research and therapy has remained unrealized, owing

to the poorly defined animal-derived matrices in which cells are grown and in particular, mostly

limited to use of Matrigel [13]. Trying to address this issue, Gjorevski et al. used polyethylene

glycol (PEG) hydrogels to define the microenvironmental parameters that govern organoid

formation [12]. That study enhanced our understanding of the mechanical regulation of intestinal

stem cells by considering the effects of matrix stiffness, which had not been examined before, as

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

167

performing controlled mechanical modulations in vivo and in Matrigel® is challenging. The

authors observed that intestinal stem cell expansion was optimal within matrices of intermediate

stiffness (1.3 kPa), but that organoid formation in low stiffness matrices was optimal (150 Pa)

[12].

In our study, applying these concepts, we propose the use of a 3D dynamic model using a

new biomaterial developed by our group: Horseradish peroxidase crosslinked silk fibroin (SF)

hydrogels [14]. The SF hydrogels easily allow the tuning of mechanical properties and identify the

most suitable stiffness to study cancer cell migration. Silk is a famous natural fiber produced by

the silkworm (Bombyx mori) cocoons and is composed of two types of protein (fibroin and

sericin), lacking the common motifs such as laminin and collagen. For this reason, silk allows the

substrates’ mechanical contribution to cell fate to be isolated better than chemically bioactive

materials, such as Matrigel. Therefore, we are able to better define and modulate the key ECM

parameters that govern colorectal cancer cell migration.

By its turn, Matrigel is used as a control, since it has already been established as 3D

substrate for modelling cancer microenvironments [15, 16]. Moreover, cell migration in this new

platform is compared to traditional 3D Boyden chamber invasion assay. Regardless of their vast

value, they do not provide tight control over the local environment, lack the ability to precisely

control the spatial organization of cells in 3D matrices, cell-cell and cell-ECM interactions.

Imaging is rather limited, disclosing the 3D microfluidic in vitro models’ physiological relevance.

The evaluation of the effect of hVCAM-1 was also assessed in both invasion models. It is now

known that VCAM-1 is involved in mediating tumor cell adhesion to vascular endothelial cells and

promoting the metastatic process [17]. Also, serum concentrations of ICAM-1 and VCAM-1 are

significantly elevated in the patients with colorectal cancer in comparison with a group of healthy

subjects [18]. One of the possible mechanisms is that the rolling cancer cells become activated

by locally released chemokines present at the surface of endothelial cells. This triggers the

activation of integrins from the cancer cells allowing their firmer adhesion to members of the Ig-

CAM family such as ICAM and VCAM-1, initiating the transendothelial migration and therefore the

extravasation process [17]. VCAM-1 is therefore thought to play a key role in the process of

malignant progression and its presence studied in terms of impact on cell migration [19].

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

168

In this work, different concentrations of SF hydrogels were produced as 3D matrices and

fine-tuned in terms of mechanical properties by means of varying the concentration of silk

solution between 1 and 3 %. The realization of a 3D CRC model was achieved by using a

commercially available microfluidic chip (Vena4™, Cellix®) suitable for intravasation and

extravasation assays comprising a 3D microwell for the testing of ECM like materials, together

with HCT-116 cancer cells. Cell migration studies using this 3D dynamic microfluidic platform in

different conditions, as well as the traditional modified Boyden chamber were carried out up to

48 hours days of culturing. Matrigel was used as control. Due to its importance during the

metastization of colorectal cancer cells, hydrogels were supplemented with the vascular adhesion

molecule hVCAM-1 to validate our model.

VI-3. MATERIALS AND METHODS

VI-3.1. Materials and reagents

Cocoons of Bombyx mori were provided by the Portuguese Association of Parents and

Friends of Mentally Disabled Citizens (Portugal).

VI-3.2. Preparation of silk fibroin hydrogels

VI-3.2.1. Preparation of silk fibroin solution

The purified silk fibroin (SF) was prepared as described previously by Yan et al. [14]. Briefly,

SF was dissolved in lithium bromide (9.3 M), followed by dialysis against distilled water for

48 hours before concentrating it using a 20 wt.% poly(ethylene glycol) solution. SF solution of

16 wt.% concentration was used for further hydrogel preparation.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

169

VI-3.2.2. Composition and in situ SF hydrogel formation

Hydrogels were prepared by mixing the concentrated SF solution with PBS at appropriate

volumes to make final 1 %, 2 % and 3 % silk solutions. To promote the SF hydrogel formation,

1 mL of SF solution was mixed with Horseradish Peroxidase (HRP) (100 µL) and hydrogen

peroxide (H2O2) (65 µL) solutions in an eppendorf in a water bath at 37 °C, as optimized

previously by our group [14]. HRP solution (0.84 mg/mL) and H2O2 (0.36 wt %) were both

prepared in PBS. For the outward cell migration (modified Boyden chamber assay), 2 % and 3 %

SF hydrogels (10 µL) were placed on top of the fluoroblock membrane (VWR, Portugal) and

allowed to crosslink for another 20 minutes. For the 3D dynamic migration assays, 2 % and 3 %

SF hydrogel discs were prepared by adding 30 μL of the mixture solutions in a 6 mm diameter

silicon mold and placed at 37 °C for crosslinking during 20 minutes. This mold size was used for

the crosslinked hydrogel to fit in the Vena4™ microwells. For the SF hydrogels supplemented with

hVCAM-1 (10 %) (Prepotech, Portugal), the solution was mixed with hVCAM-1 (20 µg/mL) and

then placed in silicon molds for crosslinking for 20 minutes. Afterwards, hydrogels were taken out

of the molds, placed in Vena4™ microwell and sealed.

VI-3.3. Preparation of Matrigel coating and hydrogel matrices: composition and in situ hydrogel

formation

Matrigel coating in outward cell migration (modified Boyden chamber) were reconstituted

from Matrigel (BD Biosciences) and diluted (250 µg/mL in 0.01 M Tris) (pH 8.0), NaCl (0.7 %)

(Laborspirit, Portugal), according to manufacturer’s instructions. Diluted Matrigel was carefully

added to the top of the membrane and incubated at 37 °C for 2 hours. Then, the remaining

liquid (coating buffer) was carefully removed from the permeable support membrane without

disturbing the layer of Matrigel. For the 3D hydrogels, Matrigel were added to the 6 mm diameter

silicon molds (30 µL) and placed at 37 °C for 30 minutes. For the hydrogels supplemented with

hVCAM-1 (10 %), Matrigel was mixed with hVCAM-1 and then placed in silicon molds for

crosslinking for 30 minutes.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

170

VI-3.3.1. Characterization of hydrogels

VI-3.3.2. Mechanical properties determination

The 1 %, 2 % and 3 % SF and Matrigel hydrogels’ storage and loss moduli were evaluated by

using an oscillatory model in a rheometer (Kinexus Prot, Malvern). For each measurement, SF

solution (1mL) was mixed with HRP (100 µL) and H2O2 (65 µL), and then the mixture (100 µL)

were transferred into 8 mm silicon molds at 37 ºC. After crosslinking, hydrogels were placed into

the rheometer for evaluation. All samples were assayed using a plate-plate geometry. The

measurements were conducted at 37 °C (n =3). Stress sweeps (0.1 Hz) were first performed to

determine LVR for all the tested conditions. Frequency sweeps (0.01-1 Hz) were then performed

within the LVR. The values of the shear storage modulus (G’) presented in Table V-1 were

obtained at a frequency of 0.1 Hz.

VI-3.4. Cell culture

HCT-116 cells (human colon cancer cell line) originally obtained from the American

Collection of Cell Cultures (ATCC, USA) were used. Cells were continuously grown in Dulbecco’s

Modified Eagle Medium (DMEM), (Sigma, Germany) supplemented with 10 % (v/v) fetal bovine

serum and 1 % penicillin and streptomycin under standard conditions (37 °C in a humidified

atmosphere containing 5 % CO2). Medium was changed twice a week and subcultures of cells

were performed when confluence reached values of ~90 %.

VI-3.5. Outward cell migration (modified Boyden chamber assay)

Cell migration was analyzed using HTS Corning FluoroBlok Cell Culture Inserts (24 well),

with an 8 µm pore size (Becton Dickenson, EUA).

First, the SF hydrogels and Matrigel coating were performed as described before. Then,

HCT-116 cells were pre-labelled with Cell Tracker Red (Invitrogen, Portugal) for 20 minutes (5

mM), and then seeded on top of the coating/hydrogels (30.000 cells per well). In the upper

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

171

chamber, either DMEM medium with 10 % v/v FBS or Serum-free medium were added (300 µL).

In the lower chamber, either DMEM medium with 10 % v/v FBS, DMEM supplemented with 10 %

v/v hVCAM-1 or with 30 % v/v FBS were added (500 µL). At different time points (0, 2, 5, 24 and

48 hours ), fluorescence intensity from the bottom was measured using a microplate

spectrofluorimeter (BioTek, Portugal) in area-scan bottom-reading mode, at excitation/emission

wavelengths of 553/570 nm. Results are presented as the increase in fluorescence in relation to

the time point 0 hours. Fluorescence images of migrating cells were collected using an inverted

fluorescence microscope (Leica, Germany). In the last time point (48 hours), cells were fixed with

formalin (10 % v/v) (Sigma, Germany) and stained for F-actin filaments of the cytoskeleton and

nuclei with Texas Red-X phalloidin (Molecular Probes, Invitrogen, USA) and with 4,6-diamidino-2-

phenylindole, dilactate (DAPI blue, Molecular Probes, USA), respectively, following supplier’s

protocol.

VI-3.6. In vitro studies under a 3D microfluidic platform

VI-3.6.1. Vena4™ Microfluidic Chip

The Vena4™ biochip from Cellix® (Irland) is constituted by 4 channels with 100 μm of

height, 800 μm of width and 2.8 cm of length. Each channel is connected to two microwells (one

in each end) where a microfluidic recirculating pump controlled by an iPod Touch (Kima™ pump)

is connected, and a microwell separated by a 8 μm pore size membrane. The sample volume of

each channel was 10 µL. The chip was fabricated in optically clear acrylic with a substrate

thickness of 500 μm, enabling the observation of cells under a brightfield, fluorescence (Leica,

Germany) and confocal microscope (Zeiss, Germany).

VI-3.6.2. Cell migration assay in 3D VenaT4™ biochips

Cell migration under a dynamic 3D microfluidic platform was conducted using VenaT4™

biochips (Cellix®, Irland). Each biochip channel was coated with fibronectin (Sigma, Germany)

(100 µg/mL) and placed in a humidified sterile petri dish and incubated at 37 °C for 1.5 hours.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

172

After the incubation period, HCT-116 cells were pre-labeled with red cell tracker CM-DiI Dye (1

µM) (Invitrogen, Portugal) and incubated at 37 ºC for 30 minutes. After, labeled cells were

seeded on Vena4™ microchannels using a standard yellow pipette (30.000 cells/channel). In

order to prevent drying, after the initial 30 minutes of incubation, complete medium DMEM was

added to the channels. The hydrogels (3 % and 2 % SF and Matrigel), previously crosslinked in

the silicon molds, were placed in the microwell and sealed. At determined time points (0 and 48

hours), the chip was observed under confocal laser scanning microscopy (Leica, Germany) to

monitor pre-labeled cells’ migration towards the hydrogel with and without hVCAM-1.

VI-3.6.3. ATP quantification

To quantify the viable cells that migrated towards the hydrogel, ATP was measured using

CellTiter-Glo Luminescent Cell Viability Assay (Promega, Portugal). For that, hydrogels were

removed from Vena4™ Biochips after 48 hours of culture and placed in 96 well plates. Then, the

analysis was performed following the manufacturer’s instruction. SF and Matrigel hydrogels were

used as controls. Briefly, CellTiter-Glo® Reagent agent was added (150 µL) to each well of the

96-well plate and incubated for 10 minutes at room temperature to lysate cells. Then, the cell

lysate (100 µL) was transfer to wells of a 96-well white opaque microtiter plate (in triplicate). This

plate was loaded into the luminometer (Perkin-Elmer, EUA). The signal intensity of the samples

was measured. Light output was given as the integral relative light units (RLUs). ATP

measurements were carried out at room temperature. An ATP standard curve was generated

using ATP solutions ranging from 5 to 1,000 cells and luminescence readings from experimental

samples were fit into this curve to generate moles of ATP.

VI-3.6.4. Live/Dead assay

The viability of cells migrated into the hydrogels was assessed using the Live/Dead assay.

Cell-laden matrices were washed three times with PBS (Sigma, Germany), then incubated (10

minutes, 37 ºC in the dark) with calcein AM (1 µM, live cells) and propidium iodide (PI, 1.5 µM,

dead cells) and washed again. Samples were imaged by confocal laser scanning microscopy.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

173

Two filters were used: for green excitation/emission (495/515 nm) and for red excitation/

emission (510/595 nm).

VI-3.7. Statistical analysis

Statistical analyses were performed using GraphPad Prism 5.0 software version 5.0a. The

non-parametric Mann–Whitney test was used to compare two groups, whereas comparison

between more than two groups was performed using the Kruskal–Wallis test followed by Dunn’s

comparison test. A value of p < 0.05 was considered statistically significant.

VI-4. RESULTS

VI-4.1. Characterization of hydrogels

VI-4.1.1. Mechanical properties determination: Rheology

The influence of polymer type and concentration of the used hydrogels was studied in terms

of their rheological behavior in order to analyze the mechanical properties of the hydrogels.

Stress sweeps (0.1 Hz) were first performed to determine the linear viscoelastic region (LVR) for

all the tested conditions. Frequency sweeps (0.01–10 Hz) were then performed within the LVR.

The values of the shear storage modulus (G’) and loss modulus (G’’) are presented in Table V-1

and were obtained at a frequency of 0.1 Hz.

Storage modulus G’ is known as the “solid-like” or elastic component of the gel, and G’’

(Loss modulus) as the “liquid-like” or viscous component. The system is considered a gel if the

value of G’ is greater than G”, which was confirmed by the results in Table V-1, where all the

values for the loss modulus are many folds smaller than the storage modulus, which was

expected for an hydrogel system that is stable during time. The higher the G’, the less it deforms

under compression and the more energy it can retain and store.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

174

Table VI-1 – Composition, rheological properties and original mesh size of hydrogels at a frequency of 0.1 Hz.

Indeed, when comparing the G’ value between SF hydrogels and Matrigel, the first is much

lower, corroborating the fact that even after crosslinking, Matrigel is difficult to handle and is not

a consistent material, which is an important feature when considering performing further studies

on the hydrogels. When comparing SF hydrogels, the G’ of 3 % SF hydrogel is much higher than

the 2 %, which is turn is much higher than the 1 % formulation. Consequently, rheology

measurements confirm that the stiffness and mechanical properties of the hydrogels increase

with increasing polymer concentrations in the case of silk fibroin, and this is due to the increase

in the number of polymer chains and crosslinking points within the system, allowing the tuning of

mechanical properties and further study of possible applications in cell migration and

tumor/spheroid formation.

VI-4.2. Outward cell migration (modified Boyden chamber assay)

This model was chosen to do a comparative study of colorectal cancer cells’ behavior

regarding cell migration. As performed in our 3D dynamic model, 2 % and 3 % SF and Matrigel

were used and the presence/absence of hVCAM-1 in order to have comparable data. Besides,

another condition was added in this experiment: a gradient of serum (from 0 % to 30 % (v/v)

FBS), generally used as chemoattractant in the modified Boyden chamber assay, to be compared

with migration observed with the gradient of hVCAM-1. As this molecule is thought to be involved

in colorectal cancer migration and metastasis, it is interesting to compare it with FBS. In the

condition described as the control, the upper and lower chamber are filled with normal DMEM

medium with no molecules’ gradient. Regarding the hydrogels, 10 µL of 2 and 3 % SF hydrogels

Type of hydrogel

Cells Incubation

(time) Storage modulus

(G’, Pa)) Loss modulus ((G'',

Pa))

1 % SF no 30 minutes 65 ± 9 8 ± 5

2 % SF no 30 minutes 488 ± 72 16 ± 4

3 % SF no 30 minutes 1136 ± 94 20 ± 3

Matrigel no 1h 45 ± 15 4 ± 0.5

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

175

were crosslinked at 37 ºC on top of the membrane. This is the minimum quantity to cover the

surface of the membrane. For Matrigel, manufactures’ protocol was followed and a dilution of

Matrigel was used to produce a thin coating. Different time points were analyzed to see the

progression of migration and corroborate the fact that this kind of phenomena occurs mainly

within the first 24 hours, as can be seen in Figure V-1 [20, 21]. Also, to prevent the quantification

of cell proliferation instead of cell migration, the experience was terminated at 48 hours

Figure VI-1 - Quantification of outward migration of HCT-116 (labeled with Red Cell Tracker in modified Boyden

chamber from 2 and 3 % SF hydrogels and Matrigel coating, in response to the presence of hVCAM-1

and 30 % (v/v) of FBS. (* indicates significant differences when comparing to Matrigel coating at

each time point; • indicates significant differences when comparing to 2 % SF at each time point; $

indicates significant differences when comparing to VCAM at each time point; ᶲ indicates significant

differences when comparing to FBS at each time point; # indicates significant differences when

comparing to time point 2 hours.

Figure V-1 shows the final results of cumulative cell migration (in terms of fluorescence

quantification) from 2 up to 48 hours. Fluorescence at 0 hours was considered our blank. The

first noticeable result is that migration phenomena starts almost immediately after the seeding on

top of the hydrogels and that the number of cells crossing through the hydrogel and the

membrane towards the bottom chamber grows until 48 hours of culturing. At 48 hours, by the

end of the experience (the same time point that is going to be analyzed in 3D model Vena4™),

and where cell migration is more evident, interesting results can be withdrawn. Both in the case

of 2 % SF and Matrigel, there is an intense cell migration in the presence of hVCAM-1,

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

176

significantly higher than in the presence of the traditional chemoattractant FBS and the control

(absence of any chemoattractant). Interestingly, cell migration to the bottom side of the

membrane increases continuously until 48 hours, as all the conditions are significantly different

when comparing to time point 2 hours. Maybe one of the most relevant observations in this

experiment is that from the beginning, especially at first time points, Matrigel allows more cell

migration than 2 and 3 % SF. Alternatively, by the last time-points, there is no substantial

differences in the fluorescence intensity between 2 % SF and Matrigel. At last, it becomes clear

that 3 % SF allows very little cell migration when comparing to 2 % SF and Matrigel.

The panel represented in Figure V-2 shows the cells that have migrated thought the

hydrogels and across the black membrane of the modified Boyden chamber. Initially, cells were

pre labeled with red cell tracker for the quantification analysis, but for better visualization, they

were later stained with DAPI (blue- nucleus) and Phalloidin (red – f-actin filaments) and observed

under inverted fluorescence microscopy. The first column of the image shows every condition,

but without cells. Therefore, the pores of the membrane are clearly visible under red fluorescence

and no cells are observable. The second column, described as control, corresponds to the

condition in which the upper and lower chambers are filled with normal DMEM medium with no

molecules’ gradient. Corroborating the results shown in the previous graphic, no cells were

observed in the bottom of membranes with 3 % SF hydrogels. In the case of 2 % SF hydrogel and

Matrigel coating, it is possible to observe that the cells have migrated and present their normal

round shape morphology.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

177

Figure VI-2 - Representative panel of outward migration of HCT-116 cells from 2 % and 3 % SF and Matrigel coating

in response to gradients of hVCAM-1 and FBS. Migrating cells attached to the lower side of the insert

membrane at 48 hours stained with DAPI (blue) and Phalloidin (Red). The first column represents

the pores of the membrane without cells. N/A denotes no cell migration.

VI-4.3. In vitro studies under a 3D microfluidic platform

VI-4.3.1. Cell migration assay in 3D VenaT4™ biochips

Although there is still a lot of uncertainty regarding the critical step in the formation of

metastatic tumors, the ability of circulating tumor cells to adhere to and transmigrate across the

endothelium at a remote site is certainly essential. In the case of this experience, the

extravasation is mimicked when adhered red labeled HCT-116 cells seeded on the microchannel

and subjected to flow perfusing media, cross the 8 µm pore membrane and migrate towards the

underlying hydrogel (from top down) and observed on the confocal microscope, as can be seen

in Figure V-3.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

178

Figure VI-3 - A) Example of Vena4™ chip with SF hydrogels placed on the microwells. B) Vena4™ placed under the

confocal microscope for migration studies. C) Retrieval of SF hydrogels out of the microfluidic chip

for further analysis

In this case, pre-crosslinked 2 and 3 % SF and Matrigel hydrogels are placed in the

microwell under the microchannels. Since literature suggests this type of migration occurs within

the first 24 hours of experience, all of our experiments lasted up to 48 hours [21, 22]. Examples

of renderings of 3D confocal stacks of all hydrogels are shown in Figure V-4. Confocal microscopy

images were taken 3 hours after cell seeding, in order to let them adhere to the microchannel in

the 3D model (data not shown), and no migration was observed in any of the tested formulations.

This means that all of the labeled cells were within the microchannel. After 48 hours of

experience, we placed the microchip in the confocal microscope and assessed cell migration.

When testing SF hydrogels as ECM mimic alternatives, in the 3 % formulation no cell migration

appeared to happen, even in the presence of hVCAM-1. However, when observing migration

experiments with 2 % formulation, results show a surprising difference and a large number of

cells had crossed the membrane and migrated towards the center of the 2 % SF hydrogel.

Moreover, in the case of 2 % SF, in the absence of hVCAM-1, we observed cells migrating about

100 - 150 µm in depth. On the other hand, in the presence of the hVCAM-1, cells could migrate

a larger distance, up to 200 µm (see white arrow in Figure V-4).

Regarding our control material, Matrigel, as depicted by the representative Figure V-4, a few

cells crossed the membrane and migrated through the micro porous membrane towards the

Matrigel. No apparent exacerbation in terms of cell migration was observed in the presence of

hVCAM-1 molecule.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

179

Summarizing the results of cell migration in the 3D microfluidic chip, we observed that when

testing SF, the softer hydrogel (2 %) allowed increasing levels of cell migration, and within this

condition, the presence of hVCAM-1 was translated in a larger distance traveled by the cells,

inside the hydrogel. However, concentrated Matrigel hydrogel, generally used in this types of

experiments, do not favor cell migration and no effect of hVCAM-1 is visible.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

180

Figure VI-4 - Representative confocal microscopy images of red-labeled HCT-116 cells’ migration towards 3 % and 2

% SF hydrogels and Matrigel, (with and without hVCAM-1) in microfluidic chip Vena4™ at 48 hours.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

181

VI-4.3.2. Live/dead assay

After assessing HCT-116 colorectal cancer cells´ migration from the microchannels towards

the hydrogels in the Vena4™ chip after 48 hours in culture, we wanted to qualitatively determine

cell viability. For that, the same experimental set up was used, but without pre-labeling the cells.

After 48 hours, the hydrogels were carefully removed from the Vena4™ microfluidic chip (Figure

V-3C) and live (green) and dead (red) cells were stained by adding Calcein-AM/Propidium Iodide

(PI) and hydrogels observed under confocal microscopy. Calcein-AM fluoresces in green in case

of living cells, as it is cleaved by esterases and produces green fluorescence, which is confined in

the living cells by an intact cytoplasmic membrane. PI is red as it is able to enter dead cells and

nuclei through the damaged membrane and binds to fragmented DNA, emitting a red

fluorescence. As observed in the representative panel in Figure V-5A, all the cells that have

migrated towards the hydrogels in all the conditions were still viable. Although this is not a

quantitative assay, it is worth noticing a higher number of (viable) colorectal cancer cells in

hydrogels supplemented with hVCAM-1, and even more pronounced in 2 % SF, corroborating the

confocal images results seen before on Figure V-4. This means that this 3D dynamic model

allows for cell viability. The hydrogels are kept hydrated, although they are not submerged in

medium, just subjected to media perfusion on top of the microchannels.

VI-4.3.3. ATP quantification

After the extravasation experiments with Vena4™, where the goal was to assess the best

formulation to observe cell migration from the microchannel to the underlying hydrogel

(confirmed through confocal microscopy), we wanted to quantitatively determine cells viability in

the hydrogels. Cells’ viability inside the hydrogels is an important parameter to determine their

suitability as ECM-like matrices. For that, the extravasation experiment was carried out during 48

hours, after which the hydrogels were retrieved from the microfluidic chip and tested for

bioluminescent-based ATP quantification. The reaction was catalyzed by the enzyme luciferase

obtained from the firefly (Photinus pyralis). The MgATP2- converts the luciferin into a form which is

capable of being catalytically oxidized by the luciferase in a high quantum yield chemiluminescent

reaction. Moreover, a calibration curve using as low as 5 cells was used to determine its

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

182

correlation with bioluminescent signal in 2D (Supplementary material). Although we cannot

directly correlate the number of cells obtained in 2D and 3D, the signal obtained when testing the

hydrogels show that the cells remained viable, corroborating the previously showed live/dead

assay results, as can be seen in Figure V-5B. Also, it was possible to see that, in agreement to

what was observed by confocal microscopy, the extravasation experiment performed with 2 % SF

shows the strongest signal, being significantly different from the 3 % formulation and Matrigel.

Figure VI-5 - A) Cell viability (live/dead assay) of HCT-116 cells that migrated in the hydrogels after 48 hours of

culture in Vena4™ microfluidic chip. B) Cell viability of 3D cultures after 48 hours of culture (2 % and

3 % SF and Matrigel hydrogels with hVCAM-1). Data is presented as mean ± stdev (n=3), (*) denotes

statistical differences (p<0.05).

VI-5. DISCUSSION

It is now more recognized than ever that the integration of microfluidic techniques together

with advanced biomaterials such as ECM mimicking tunable hydrogels can supply a unique

platform to develop 3D organized constructs mimicking the in vivo tumor microenvironment [23,

24]. Concerning this, it was our goal to take advantage of our group’s great expertise in

biomaterials and test SF hydrogels in the context of colorectal cancer metastasis. Silk is an

excellent candidate for cancer related studies due to its biocompatibility and highly tunable

properties [14]. To the best of our knowledge, there has been lacking specific attempts to use

tunable silk hydrogels as 3D ECM like materials to push forward the field of colorectal cancer

metastasis research, combined with a confocal compatible microfluidic platform.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

183

This innovative work aims at using a dynamic 3D microfluidic platform to test several silk

hydrogel formulations for HCT-116 cell migration (extravasation). Also, an extensive comparative

study between the traditional modified Boyden chamber assay and our 3D microfluidic chip to

assess colorectal cancer cell invasion, using 2 and 3 % SF hydrogels and Matrigel as a control

group was performed. Although 1 % SF hydrogels were produced and characterized in terms of

rheology, they were not tested for cell migration, as the 2 % silk formulation was the highest

formulation to allow HCT-116 cancer cell migration in a dynamic environment, and robust

enough to retrieve for further analysis such as ATP quantification and live/dead (Figure V-3).

Moreover, the influence of hVCAM-1, a molecule widely present in colorectal cancer metastasis

processes is also assessed in order to validate our 3D dynamic model [17].

An innovative aspect of our study is that in our proposed 3D model, unlike most models,

which focus on seeding cells on hydrogel sheets as well as encapsulating them within the 3D

matrices, we seed cells on a microchannel, allow them to adhere, and observe their migration

through a microporous 8 µm membrane and towards the hydrogel [25]. Therefore, this

extravasation model represents a more realistic scenario of the extravasation process [26].

Literature suggests metastasis assays, being intravasation or extravasation, generally occurs

within the first 24 hours of tumor cell contact with the endothelium, therefore we decided to

study this phenomena within the first 48 hours [27].

Based on the widely used modified Boyden chamber, we determined cell invasion by

quantifying the outward cell migration of fluorescent labeled cells. In this assay, a layer of ECM

like matrix is deposited on the porous membrane. Although collagen and laminin are also used,

this layer is typically a Matrigel coating [28]. Results of migration show that migration rates are

similar between 2 % SF and Matrigel at 48 hours, with a significant increase when in the

presence of hVCAM-1 in both cases. This result is relevant because it shows differences between

hVCAM-1 and FBS gradients, being the latest, almost always used as chemoattractant [9]. It

exposes the real influence of mimicking the in vivo microenvironment cues when studying cancer

cell migration. Regarding the 3 % SF conditions, it becomes clear that the matrix does not favor

cell migration in the Boyden chamber assay. In addition, in the first hours of the assay, results

were striking in terms of the difference between cancer cell migration in Matrigel and 2 % SF,

showing that Matrigel coating allows significantly more migration than 2 % SF.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

184

When moving to our newly developed 3D microfluidic model, results are outstandingly

different. As can be seen in Figure V-4, after 48 hours of dynamic culture, when analyzing the

tissue engineered 3D model in the confocal microscope, no noticeable cell migration was

perceived, and no observable effect was attributed to the presence of hVCAM-1 in the 3 % SF

hydrogel. When moving to the 2 % formulation, we observed it allowed for a large number of cells

to move from the microfluidic channel, through the microporous membrane and into the

hydrogel. In the case of 2 % SF hydrogel crosslinked in the presence of 10 % hVCAM-1, it was

translated in a greater distance traveled by the cells. More quantitative assays are needed to

corroborate this information, but the representative confocal images leave no doubt about this

change of behavior and add evidence on the association of hVCAM-1 with malignant potential in

colorectal cancer. On the other hand, when comparing the observations in the two tested

scenarios (3D static (modified Boyden chamber assay) and 3D dynamic (our proposed

microfluidic model)), a major difference is detected in cell behavior in the case of our control

material: Matrigel. Whereas in the Boyden chamber cells migrate in large numbers from the first

hours in the Matrigel coating (Figure V-2), in the dynamic 3D model, this does not happen (Figure

V-4). These findings came as a surprise when analyzing the rheology results of the hydrogels. In

fact, considering the data from rheology analysis it was expected that 2 % SF, (G’ value of 488 ±

72 Pa) allowed less cell migration than Matrigel (G’ value of 45 ± 15 Pa), since it is stiffer.

Nevertheless, the opposite trend was observed, with more cell migration in the stiffer 2 % silk

hydrogel than the softer Matrigel. This phenomena could be explained by the different

composition of the hydrogels. In this sense, silk hydrogels has no binding motifs (e.g. RGD

peptide), but Matrigel has natural mammalian binding motifs. It has already been discussed and

reported in the literature a possible biphasic role of Matrigel that can partially explain these

phenomena: at low concentration, such as coating, Matrigel facilitates migration, most probably

by providing a supportive and growth factor retaining environment. At high concentration,

Matrigel slows down migration, possibly due excessive attachment [22]. In the Boyden chamber

assay, following suppliers instructions, we used a Matrigel coating by diluting Matrigel to 250

µg/mL, nearly 20 fold less concentrated then the hydrogels used in our 3D dynamic model.

On the other hand, silk hydrogels obtained from Bombyx mori are not completely deprived of

mammalian binding motifs. Silk proteins are comprised of a main heavy chain, which can be

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

185

considered a hydrophobic protein with co-block design [29]. Hydrophilic segments are involved in

the self-assembling process, resulting in changes in water content and regulation of the

mechanical properties of the final material, which can explain cell migration in these matrices

[29]. In addition, two short aminoacid sequences that are RGD-like are present in the N-terminus

segment of heavy-chain, VTTDSDGNE and NINDFDED, as well as the recognized fibroblast

integrin [29].

In fact, Friedl et al. described two different migration movements, mesenchymal and

amoeboid migration [30]. In the first, mesenchymal migration, cells exhibit high attachment and

cytoskeletal contractility, which allow cells to migrate. But in the second, amoeboid migration,

cells present poor attachment and a lack of stress fibers, which results in a round morphology.

For so, they develop a strategy that consists in the formation of blebs and the use of propulsive

forces to migrate. In this sense, the 2 % SF matrices possess mechanical properties that can

allow amoeboid movements [30]. In the case of 3 % matrices, similar results were observed

when comparing the Boyden chamber assay to the proposed 3D dynamic model. The resistance

to cell migration in the 3 % SF formulation, even in the presence of chemoattractant, can be

explained by matrix stiffness. By its turn, a G’ value of 1136 ± 94 Pa of the 3 % silk formulation is

simply too stiff to allow cell migration through the polymer’s fibers, as confirmed by Bott et al.,

who revealed that in spite of matrix sensitivity to proteases (e.g. MMP) and the presence of cell-

integrin binding sites, at high stiffness (G′ >1200 Pa) the matrix acts as a barrier for cells

cultured in 3D to migrate [31]. As a matter of fact, these results are in alignment with those

observed by Gjoresvki et al., showing that in the case of PEG hydrogels, a higher matrix stiffness

of 1300 Pa is suitable for cell expansion, but organoid development is optimal in a later stage, in

gels with a lower stiffness of around 100 Pa, which can be achieved by the presence of

metalloproteinases [12]. In our case, we are interested in studying cell migration, a different

phenomenon than cell expansion and organoid formation, where the intermediate stiffness

seems to work best. In a further stage, by using longer periods of culture in dynamic conditions,

we aim at creating a tumoroid by modulating silk fibroin hydrogels mechanical properties to

achieve lower stiffness, such as the 1 % SF formulation (64 ± 9 Pa, Table V-1).

It is important to bear in mind that the effects of ECM properties on the normal epithelial

morphogenetic program are fundamentally different from those reported for cancer cells in

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

186

synthetic gels, most likely reflecting central differences between developmental morphogenesis

and tumorigenesis [32]. Matrix stiffness is therefore a crucial biophysical aspect of the tumor

microenvironment and, consequently, has been profoundly studied in collagen, polyacrylamide,

and Matrigel hydrogels, but not in silk [33]. Therefore, the proposed 3D dynamic model offers an

opportunity to better understand the role of mechanical properties of the matrix environment,

assessing specific effects of matrix stiffness on colorectal cancer progression [33, 34].

Taken together, these facts can explain the differences we found between the traditional

static and dynamic models and disclose the importance of using relevant in vitro models.

After our study, we believe we are facing a question of developing a microenvironment

mimic material, ultimately choosing between bio-based vs polymer-based ECM like materials.

Although there is undoubtedly a growing number of studies based on polymers, Matrigel, a cell

secreted protein mixture, has been considered the standard material in terms of cell migration.

However, since further development of chip-based 3D cell culture in cancer research will be

largely dependent on the improvement of biomaterials that emulate the ECM and the capacity to

scale-up these complex technologies, Matrigel may not represent the best choice, for several

reasons: Its composition is undefined and cannot be fully controlled, changing from batch to

batch, with a complex mixture of components (unknown amounts of growth factors and proteins)

that may play a role in cell behavior. In terms of practical use, it’s not as user friendly as silk, due

to its gelling temperature and stiffness/consistency to perform further tests on the hydrogels [35,

36]. Moreover, it won’t allow long term experiments, very much needed to understand the real

process of tumor and metastasis formation. Development of integrating TE approaches and

microfluidics into easy-to-use, scalable, reproducible and cost-effective systems will be the key to

their success and future translation to the market [37].

The Boyden chamber assay, exploiting a chemokine gradient between upper and lower

chambers to drive cell invasion and migration, can be readily adapted to model specific

extracellular matrix (ECM) chemistries by simple coating procedures and is undoubtedly a

precious tool. However, a comprehensive assessment of three-dimensional tumor cell invasion or

migratory mechanisms using histological techniques is difficult to achieve in these systems, with

metrics restricted to an end-point summation of cell numbers in the lower chamber. Controlling

the spatial distribution of cells on extracellular proteins at the 2D interface is challenging and we

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

187

anticipate that with scientific research development, biomaterial platforms with controlled x, y

and z internal dimensions will prove a valuable improvement. Additionally, the developed 3D

dynamic model presents several advantages such as the microchannels in contact with the 3D

microenvironment, flow, the real-time tracking of cells, a fine balance between complexity and

experimental control and the ability to isolate the effect of variables. Horseradish crosslinked

hydrogels represent a suitable ECM like material, with the advantage of being easily tunable in

order to best match the desired application. However, this model still lacks some aspects in

order to represent the real complexity of tumor microenvironment. For instance, the use of

primary tumor cells (making it a patient specific assay when testing drugs) and a confluent

monolayer of endothelial cells along the microchannel will add the very much needed complexity

of the cell–cell junctions between endothelial cells and the ECM that they produce, being this our

next goal.

VI-6. CONCLUSIONS

3D In vitro dynamic models allow scientists to study aspects of the tumor microenvironment

using specific extracellular matrices and cell types. Controlling and understanding the effect of

the various components of these models pushes the field forward and enables investigation of

interactions within the tumor microenvironment, as well as the response to stimuli such as

chemoattractants and chemotherapeutics. However, microfluidic models need standardization

and reproducible formats suitable for high-throughput applications. If cell migration studies are to

be used to test drug efficacy in this context, 3D models have to be reliable with regard to

fabrication and incorporation of the most adequate ECM like material to reduce arbitrary cell

migration patterns.

We explored for the first time silk hydrogels as ECM like materials to mimic tumor

microenvironment and to study colorectal cancer migration, compared to a control group:

Matrigel. Moreover, after comparing the proposed 3D dynamic microfluidic model to the

traditional Boyden chamber assay, fundamental differences in terms of colorectal cancer cell

migration were shown. Ultimately, we developed a tunable Silk hydrogel for the development of a

CRC Extravasation 3D Model on a Chip. This model represents a valuable tool to better

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

188

understand cell migration and tumorigenesis processes, consisting in a proof of concept, where

different materials can be assessed as ECM like materials, as well as the influence of

chemoattractants.

VI-7. ACKNOWLEDGMENTS

This article is a result of the project FROnTHERA (NORTE-01-0145-FEDER-000023),

supported by Norte Portugal Regional Operational Programme (NORTE 2020), under the

PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund

(ERDF). Mariana Carvalho acknowledges her PhD scholarship NORTE-08-5369-FSE-000044,

funded by Programa Operacional Regional do Norte, Fundo Social Europeu, Norte2020

TERM&SC. F.R. Maia acknowledges Portuguese Foundation for Science and Technology (FCT) for

her Postdoc scholarship (SFRH/BPD/117492/2016). Sílvia Vieira is also greatly acknowledge

FCT for her PhD scholarship (SFRH/BD/102710/2014). J. M. Oliveira thanks FCT for his

distinction attributed under the FCT Investigator program (IF/00423/2012 and

IF/01285/2015).

VI-8. REFERENCES

1. Haggar, F.A. and R.P. Boushey, Colorectal Cancer Epidemiology: Incidence, Mortality, Survival, and Risk Factors. Clinics in Colon and Rectal Surgery, 2009. 22(4): p. 191-197. 2. Carvalho, M.R., et al., Significance of Tissue-engineered 3D in-vitro models for cancer research and drug screening. Trends Biotechnol, 2015. 3. Mak, I.W., N. Evaniew, and M. Ghert, Lost in translation: animal models and clinical trials in cancer treatment. Am J Transl Res, 2014. 6(2): p. 114-8. 4. Katt, M.E., et al., In Vitro Tumor Models: Advantages, Disadvantages, Variables, and Selecting the Right Platform. Frontiers in Bioengineering and Biotechnology, 2016. 4: p. 12. 5. Carvalho, M.R., et al., Evaluating Biomaterial- and Microfluidic-Based 3D Tumor Models. Trends in Biotechnology. 33(11): p. 667-678. 6. Wlodkowic, D. and Z. Darzynkiewicz, Microfluidics: Emerging prospects for anti-cancer drug screening. World J Clin Oncol, 2010. 1(1): p. 18-23. 7. Majedi, F.S., et al., On-Chip Fabrication of Paclitaxel-Loaded Chitosan Nanoparticles for Cancer Therapeutics. Advanced Functional Materials, 2014. 24(4): p. 432-441. 8. Maia, F.R., et al., Matrix-driven formation of mesenchymal stem cell-extracellular matrix microtissues on soft alginate hydrogels. Acta Biomater, 2014. 10(7): p. 3197-208.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

189

9. Justus, C.R., et al., In vitro Cell Migration and Invasion Assays. Journal of Visualized Experiments : JoVE, 2014(88): p. 51046. 10. Liu, J., et al., Soft fibrin gels promote selection and growth of tumorigenic cells. Nat Mater, 2012. 11(8): p. 734-41. 11. Fang, X., et al., Novel 3D co-culture model for epithelial-stromal cells interaction in prostate cancer. PLoS One, 2013. 8(9): p. e75187. 12. Gjorevski, N., et al., Designer matrices for intestinal stem cell and organoid culture. Nature, 2016. 539(7630): p. 560-564. 13. Fatehullah, A., S.H. Tan, and N. Barker, Organoids as an in vitro model of human development and disease. Nat Cell Biol, 2016. 18(3): p. 246-54. 14. Yan, L.P., et al., Tumor Growth Suppression Induced by Biomimetic Silk Fibroin Hydrogels. Sci Rep, 2016. 6: p. 31037. 15. Benton, G., et al., Matrigel: From discovery and ECM mimicry to assays and models for cancer research. Advanced Drug Delivery Reviews, 2014. 79–80: p. 3-18. 16. Bielecka, Z.F., et al., Three-dimensional cell culture model utilization in cancer stem cell research. Biol Rev Camb Philos Soc, 2016. 17. Gout, S. and J. Huot, Role of Cancer Microenvironment in Metastasis: Focus on Colon Cancer. Cancer Microenvironment, 2008. 1(1): p. 69-83. 18. Schlesinger, M. and G. Bendas, Vascular cell adhesion molecule-1 (VCAM-1)—An increasing insight into its role in tumorigenicity and metastasis. International Journal of Cancer, 2015. 136(11): p. 2504-2514. 19. Velikova, G., et al., Serum concentrations of soluble adhesion molecules in patients with colorectal cancer. British Journal of Cancer, 1998. 77(11): p. 1857-1863. 20. Joyce, J.A. and J.W. Pollard, Microenvironmental regulation of metastasis. Nat Rev Cancer, 2009. 9(4): p. 239-52. 21. Bersini, S., et al., A microfluidic 3D in vitro model for specificity of breast cancer metastasis to bone. Biomaterials, 2014. 35(8): p. 2454-61. 22. Anguiano, M., et al., Characterization of three-dimensional cancer cell migration in mixed collagen-Matrigel scaffolds using microfluidics and image analysis. PLOS ONE, 2017. 12(2): p. e0171417. 23. Alemany-Ribes, M. and C.E. Semino, Bioengineering 3D environments for cancer models. Adv Drug Deliv Rev, 2014. 79-80: p. 40-9. 24. Peela, N., et al., A three dimensional micropatterned tumor model for breast cancer cell migration studies. Biomaterials, 2016. 81: p. 72-83. 25. Kraning-Rush, C.M. and C.A. Reinhart-King, Controlling matrix stiffness and topography for the study of tumor cell migration. Cell Adh Migr, 2012. 6(3): p. 274-9. 26. Chen, M.B., et al., On-chip human microvasculature assay for visualization and quantification of tumor cell extravasation dynamics. Nat Protoc, 2017. 12(5): p. 865-880. 27. Jeon, J.S., et al., In vitro model of tumor cell extravasation. PLoS One, 2013. 8(2): p. e56910. 28. Marshall, J., Transwell® Invasion Assays, in Cell Migration: Developmental Methods and Protocols, C.M. Wells and M. Parsons, Editors. 2011, Humana Press: Totowa, NJ. p. 97-110. 29. Floren, M., C. Migliaresi, and A. Motta, Processing Techniques and Applications of Silk Hydrogels in Bioengineering. Journal of Functional Biomaterials, 2016. 7(3): p. 26.

Chapter VI–Tuning Enzymatically-crosslinked Silk fibroin Hydrogel Properties for the Development of a Colorectal cancer Extravasation 3D Model on a Chip

190

30. Friedl, P. and K. Wolf, Plasticity of cell migration: a multiscale tuning model. The Journal of Cell Biology, 2010. 188(1): p. 11-19. 31. Bott, K., et al., The effect of matrix characteristics on fibroblast proliferation in 3D gels. Biomaterials, 2010. 31(32): p. 8454-64. 32. Enemchukwu, N.O., et al., Synthetic matrices reveal contributions of ECM biophysical and biochemical properties to epithelial morphogenesis. J Cell Biol, 2016. 212(1): p. 113-24. 33. Tilghman, R.W., et al., Matrix Rigidity Regulates Cancer Cell Growth by Modulating Cellular Metabolism and Protein Synthesis. PLoS ONE, 2012. 7(5): p. e37231. 34. Fischer, R.S., et al., Stiffness-controlled three-dimensional extracellular matrices for high-resolution imaging of cell behavior. Nature protocols, 2012. 7(11): p. 10.1038/nprot.2012.127. 35. Fischbach, C., et al., Engineering tumors with 3D scaffolds. Nat Methods, 2007. 4(10): p. 855-60. 36. Asghar, W., et al., Engineering cancer microenvironments for in vitro 3-D tumor models. Materials Today, 2015. 18(10): p. 539-553. 37. Wong, A.P., et al., Partitioning microfluidic channels with hydrogel to construct tunable 3-D cellular microenvironments. Biomaterials, 2008. 29(12): p. 1853-61.

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

191

Chapter VII

Colorectal Tumor-on-a-chip System:

A 3D Tool for Precision Onco-

nanomedicine

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

192

Chapter VII

Chapter VII - Colorectal Tumor-on-a-chip System: A 3D Tool

for Precision Onco-nanomedicine1

ABSTRACT

Awareness that over-simplistic traditional 2-dimensional (2D) in vitro and non-representative

animal models may not completely emulate the 3D hierarchical complexity of tissues and organs

is on the rise. Therefore, posterior translation into successful clinical application is compromised,

with a critical negative impact in cancer treatment. To address this dearth, on-chip biomimetic

microenvironments powered by microfluidic technologies are being developed to better capture

the complexity of in vivo pathophysiology. Herein, we describe a “tumor-on-a-chip” model for

assessment of precision nanomedicine delivery on which we validate the efficacy of drug-loaded

nanoparticles in a gradient fashion. The model validation was performed by viability studies

integrated with live imaging to confirm the dose-response effect of cells exposed to the

CMCht/PAMAM nanoparticles gradient, with 40 % cell dead after 5 days of Gemcitabine

exposure. Besides the advantages of a 3D biomimetic model supported by microfluidic culture,

this platform also enables the analysis at the gene expression level. We observed a down-

regulation of all the studied genes (MMP-1, Caspase-3 and Ki-67) after 5 days of culture. Overall,

this tumor-on chip model clearly represents an important development in the use of precision

nanomedicine towards personalized treatment.

Keywods: Microfluidics; 3D tumor model; Colorectal cancer; Nanoparticles.

1This chapter is based on the following publication:

M.R.Carvalho, D. Barata, L. M.Teixeira, S. Giselbrecht, R. L. Reis, J. M. Oliveira, R. Truckenmüller, P. Habibovic.

“Colorectal tumor-on-a-chip system: a 3D tool for precision onco-nanomedicine”. Science Advances, 2019, 5:

eaaw1317. doi: 10.1126/sciadv.aaw1317.

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

193

VII-1. GRAPHICAL ABSTRACT

VII-2. INTRODUCTION

Colorectal cancer (CRC) is now the third leading cause of cancer-related death in women

and the second leading cause in men, mainly as result of high incidence of cancer metastasis

and chemotherapeutic treatments inefficiency [1]. Concerning cancer research, currently

available 2D culture systems and animal models used to study the efficacy of chemotherapy do

not reproduce the tumor physiology in the human body. As a deleterious consequence, over 80 %

of drug candidates of all cancer drugs fail during phase II and phase III clinical trials [2].

The need for a more reliable and predictable screening approach to better assess drug

responses in preclinical studies in vitro has led to the development of novel methodologies for

cell culture, employing e.g. microfluidics [2, 3]. Over the past decade, the expansion in the field

of microfluidics has helped to establish a new set of standards in the study of basic biological

phenomena, especially helpful in the study of pathogenesis and drug development [4].

Microfluidics is by definition the science and technology of systems that process and manipulate

small (10−9–10−18 L) amounts of fluids by using channels with dimensions from tens to hundreds

of micrometers [5]. Here, microfluidic technology offers new opportunities for cell-based sensors

and multifunctional platforms for biochemical and biomedical functions under physiologically

relevant conditions [6, 7].

In fact, microfluidic technologies enable precise control over small fluid/liquid volumes,

compartmentalization and combinatorial factors/materials in a single device, allowing cells to co-

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

194

exist in a 3D complex microenvironment [8]. Such kind of platforms have proven useful in

unraveling key cellular behavioral aspects of cancer metastases and enabling precise and

representative drug screening [6, 9]. In addition, microfluidics have also been successful in the

generation of vasculature-like structures, with special emphasis on the vital association between

vasculature/angiogenesis and cancer [3, 10-12]. These are useful for revealing the mechanisms

in the development of vasculatures, but also as experimental platforms for vascular disease

models and drug screening. Another particular feature of microfluidics, which will be employed in

this study, is the ability to generate well-controlled gradients of molecules in solution. Gradients of

molecules can be generated by advection, being the mass transport by the fluids bulk motion

(flow-based gradient generation), or by diffusion, being the spreading of unequally distributed

molecules by biased random walk induced by Brownian motion (diffusion-based gradient

generation) [13]. The importance of gradient generation relies on the fact that many cell

processes work by recognizing the directional information present in gradients, including

processes such as chemotaxis, angiogenesis, and diseases such as cancer [14]. In the in vivo

scenario, chemical microenvironment includes nutrients and oxygen, delivered to cell tissues

through a net of vasculature that create important biomolecular gradients within the tissue. The

successful development of 3D micro-engineered gradients set the stage for physiologically

relevant in vitro models, comprising for instance oxygen, chemokines and drug gradients [6, 13,

15]. Another important aspect of our model is the microvasculature. It defines the biological and

physical characteristics of the microenvironment within tissues and plays a role in the initiation

and progression of many pathologies, including cancer [16]. Microfluidics have also been used in

the generation of vasculature-like structures, with special emphasis on the association between

vasculature/angiogenesis and cancer [3, 10-12]. In the case of CRC, endothelial cells lining the

microvasculature are known to play a critical “gatekeeper” role, which is why we developed our

microvasculature model using Human Colonic Microvascular Endothelial cells (HCoMECs) [17].

Regarding CRC treatment, Gemcitabine (2′,2′-difluoro-2′-deoxycytidine) (from now on

designated as GEM) is now being tested as treatment for patients with advanced colorectal

cancer [18, 19]. However, because of its low molecular weight and high solubility in water, the

clinical benefits of GEM are hampered by its short plasma half-life and relative low concentration

around tumor sites [20, 21]. Therefore, a search for novel therapeutic strategies is imperative,

such as the use of nanoparticles [20].

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

195

Our goal was to develop a complex 3D microfluidic chip-based in vitro model that emulates

the human colorectal tumor microenvironment. This model enables the reconstitution of

physiological functions of microvascular tissue, which is instrumental to assess the efficiency of

delivery of anti-cancer drug-loaded nanoparticles through a dynamic controllable gradient in a

core compartment of the microfluidic chip. Such gradient is generated and maintained by

perfused microvasculature-mimicking side channels. Together, these results demonstrate that

this 3D platform is suitable for efficacy/toxicity screening in a more physiological environment

than previously possible.

VII-3. MATERIALS AND METHODS

VII-3.1. Microfluidic chip design and fabrication

The microfluidic device composed by three main compartments was designed using

computer assisted design (CAD) on a photolithography mask, which consisted in: a central

chamber for an ECM like hydrogel (5 mm diameter) and a pair of perfusable channels flanking

the center chamber (connected to the chamber with a range of pillars that allow the generation of

stable gradients by diffusion).

The microfluidic device in poly(dimethylsiloxane) (PDMS, Down Corning) elastomer was

fabricated by replica molding from a SU-8 (Microchem)/silicon master, based on a 10:1 ratio to

curing agent. The master mold was produced by UV-lithography. After pouring PDMS on mold,

the mixture was degassed in a vacuum chamber for air bubble removal, and cured at 80 °C for

1 hour, in the oven. The devices were then cut out by a razor blade, the fluidic connection ports

punched, and bonding to a glass slide done after oxygen plasma of both surfaces and conformal

contact (Plasma cleaner/sterilizer Harrick®, high RF level, 4 mbar, 120 seconds). Assembly

quality control and characterization was done by optical 3D profilometry (Keyence VK-X250,

Germany).

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

196

VII-3.2. Characterization of the microfluidic device

This microfluidic device functions as a multiphase micro-bioreactor, which was built based

on three major compartments: a central chamber seeded with cells embedded in a hydrogel

matrix (Matrigel used as model gel), being sided by a pair of channels perfused with culture

medium. The central chamber and the lateral channels are separated by small pillars to allow for

hydrogel loading of the chamber retained before the channels. Quality check was done by optical

microscopy.

VII-3.3. Cell Culture

HCT-116 cells (human colon cancer cell line) were originally obtained from the American

Collection of Cell Cultures (ATCC, USA). Cells were cultured in Dulbecco’s Modified Eagle

Medium (DMEM, Gibco, Invitrogen) supplemented with 10 % (v/v) fetal bovine serum (FBS) and 1

% (v/v) penicillin and streptomycin (pen/strep) solution under standard conditions (37 °C in a

humidified atmosphere, 5 % CO2). HCoMECs (Human Colonic Microvascular Endothelial cells)

were obtained from Innoprot and cultured in Endothelial Cell Medium supplemented with 5 %

(v/v) FBS, 1 % (v/v) of Endothelial Cell Growth Supplement (ECGS) and 1 % (v/v) of pen/strep

solution. Medium was replaced every 3 days. Subculture of cells was performed before 90 % of

confluence using 0.25 % (v/v) trypsin/EDTA (GIBCO) for 5 minutes at 37 °C.

VII-3.4. Cell seeding on the chip

HCT-116 cancer cells were re-suspended in Matrigel supplemented with VEFG (100 ng/mL)

(R&D Systems) in a density of 10x106 cells/mL, injected in the central chamber of the

microfluidic device using a syringe and let incubate for 15 min at 37 ºC for crosslinking. After

this, the lateral channels were coated using a 10 % Matrigel solution and incubated for 30

minutes at 37 ºC. HCoMECs in the same density (10x106 cells/mL) were seeded on the lateral

channels using a pipette and left to adhere for 6 hours. Then, the cells were exposed to a

perfusion flow regime at a flow rate of 8 μL/h for the duration of the experiment. At days 1 and 5

days, the cell culture was evaluated using fluorescence microscopy.

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

197

VII-3.5. Validation of the co-culture by fluorescence imaging

Prior to seeding in the microfluidic device, HCT-116 and HCoMEC cells were made

fluorescent by means of CellTracer™ Far Reed and Yellow (Life Technologies), respectively. These

dyes were designed to be easily internalized, passing through cell membrane into cytoplasm. The

desired concentration of cells trypsinized, centrifuged and resuspended in dye solution/PBS (at a

concentration of 10 µM). The suspension was incubated for 20 minutes at 37 ºC (in the dark),

and then diluted in five times the original volume of medium, let incubation for 5 minutes and

centrifuged. Finally, the pellet was resuspended in fresh medium into a final concentration and

cell suspension seeded. Cells in microfluidic culture were observed under fluorescent microscope

at day 1 and 5. In addition, afterwards, all microfluidic culture was fixed for the further analysis of

the microfluidic co-culture model. Endothelial cells seeded on the lateral channels, cells were

stained with DAPI/Phalloidin. Cells were fixed with 4 % paraformaldehyde in PBS for 30 minutes

at room temperature (RT), and washed again with PBS. Permeabilization of cells was done with

0.1 % Triton X-100 (Sigma-Aldrich) for 15 minutes at RT, followed by washing with PBS. High-

affinity filamentous actin probe Alexa Fluor 594® Phalloidin (Invitrogen™, 1:80) was injected

through the channels and let incubate for 30 minutes at RT, washed with PBS, and then

incubated with 4’,6-diamidino-2-phenylindole (Invitrogen™, DAPI, 1:100) diluted in PBS for 10

minutes at RT.

VII-3.6. Endothelial invasion into microfluidic core and quantification

During cell culture, bright field images were taken with inverted microscope to analyze and

quantify in which extent endothelial cells remain on the lateral microchannels or respond to

Vascular Endothelial Growth Factor (VEGF) present in the Matrigel in the central chamber. Three

images of each time point (day 0, day 2, day 3 and day 5) were analyzed. The length was

measured using Image J (Fiji), from the outer limit of the microchannels/hydrogel interface, until

the furthest point of the elongated cell(s).

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

198

VII-3.7. CMCht/PAMAM dendrimer nanoparticles

VII-3.7.1. Chemical synthesis of CMCht/PAMAM dendrimer nanoparticles

Dendrimer nanoparticles were surface modified as previously described by Oliveira et al.

[22]. CMCht with a degree of 80 % deacetylation and 47 % substitution of was synthesized by a

chemical modification route of chitin (Sigma, Germany) as described by Chen et al. [23].

Starburst PAMAM carboxylic acid terminated dendrimers, hereafter designated as PAMAM-

CT (G1.5, 20 %wt. methanolic solution) with an ethylenediamine core were purchased (Sigma-

Aldrich). CMCht/PAMAM dendrimer nanoparticles were prepared in a step-wise manner as

follows: increase the generation of the PAMAM-CT (G1.5), obtaining a PAMAM methyl ester

terminated dendrimer, reaction between PAMAM and CMCht (the reaction occurs by a

condensation reaction between the methyl ester and amine groups), and conversion of methyl

ester groups that do not react into carboxylic groups in the CMCht/PAMAM dendrimer, followed

by precipitation.

VII-3.7.2. Labeling of CMCht/PAMAM dendrimer nanoparticles with FITC

The conjugates of CMCht/PAMAM-FITC were obtained by covalently bonding the amine

group of CMCht and the isothyocianate group of FITC (10 mg/mL FITC) (Sigma, Germany) in

anhydrous dimethyl sulfoxide (DMSO, Norconcessus) creating a thiourea bond. First, a 10

mg/mL CMCht/PAMAM dendrimer nanoparticle solution was prepared in a carbonate-

bicarbonate coupled buffer of pH 9.2. Then, a solution of FITC/DMSO was added under agitation

and kept in the dark at 4 °C for 8 hours. The FITC-labeled CMCht/PAMAM dendrimer

nanoparticle solution was then dialyzed against ultra-pure water. The final product was obtained

after freeze-drying.

VII-3.7.3. Gemcitabine incorporation into CMCht/PAMAM dendrimer nanoparticles

GEM was incorporated in the nanoparticles by mixing an aqueous solution of

CMCht/PAMAM dendrimer nanoparticles with an aqueous GEM solution with a final

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

199

concentration of 0.5 mM. Saturated sodium carbonate solution (Na2CO3, Aldrich, Germany) and

acetone (Pronalab, Portugal) were then added to the mixture. The resulting precipitates were

dialyzed against ultrapure water. GEM-loaded CMCht/ PAMAM dendrimer nanoparticles were

obtained by freezing the final solution at -80 ºC and freeze-drying (Telstar-Cryodos-80, Spain) the

samples for 7 days.

VII-3.7.4. Encapsulation efficiency, drug loading efficiency and release profile studies

For assessing encapsulation efficiency, CMCht/PAMAM (50 mg) was dissolved in 50 mL

deionized water. GEM, (7 mg) in 2 mL deionized water was added drop-wise to the dendrimer

solution and stirred for 30 minutes. Precipitation of the dendrimers and consequent

encapsulation of the drug was induced as described before. Samples were taken before and after

precipitation, centrifuged, analyzed by UV-Vis spectrophotometer at 275 nm, in a Hellma Quartz

Plate, and compared to a calibration curve.

For evaluating drug-loading efficiency, pre-weighted amount of GEM-loaded CMCht/PAMAM

(50 mg) was added to distilled water and sonicated to extract GEM. The extracted GEM

concentration was analyzed by UV-Vis spectrophotometer at 270 nm and compared to a

calibration curve. The absorbance was used to calculate the percentage of GEM encapsulated.

The following Equations VII-1 and VII-2 were used:

Equation VII-1 – Determination of encapsulation efficiency.

Equation-VII-2 – Determination of drug loading efficiency.

GEM release from nanoparticles was measured by UV spectrophotometry (Shimadzu) at 275

nm, in a Hellma Quartz Plate, after dissolution of 50 mg of GEM-loaded CMCht/PAMAM

dendrimer nanoparticles in 5 mL of phosphate buffer saline solution (PBS) with sodium azide

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

200

0.01 % (Sigma). The in vitro release was performed at 37 ºC under stirring at 60 rpm between 0

hours and 5 days timepoints.

VII-3.7.5. Characterization of the CMCht/PAMAM dendrimer nanoparticles by transmission

electron microscope (TEM)

The nanoparticles morphology was investigated by transmission electron microscopy (TEM;

Philips CM-12, FEI Company, The Netherlands, equipped with a MEGA VIEW-II DOCU camera

and Image Software Analyzer SIS NT DOCU) after staining with 2 % of phosphotungstic acid and

fixation on copper grids.

VII-3.8. Gradients generation and characterization inside the microfluidic chip

For the generation of a stable gradient through the Matrigel core compartment, a source-

sink mechanism was created, with one solution of 0.5 mg/mL FITC-labeled nanoparticles in PBS

perfused through one side channel and PBS was perfused through other, using a Nexus 3000

syringe pump (Chemix). Syringes of 3 mL syringes were used to introduce PBS with or without

the nanoparticles at a rate of 8 µL/hour. The experiment was performed in a Nikon Eclipse Ti-E

fluorescence microscope (Nikon Instruments, Japan) for time lapse imaging of the gradient

formation.

Profile and stability of the dynamic gradient was analyzed using NIS-Elements AR (Nikon)

software. The chips were analyzed at different time points between 0 and 12 hours with

continuous perfusion. Built-in “plot profile” module was used for assessing the gradient intensity

and stability over time. The intensity profile corresponds to distance versus intensity. A 3D

projection was also reconstructed using software.

VII-3.9. Gemcitabine loaded CMCht/PAMAM dendrimer nanoparticles gradient

After the establishment of the co-culture model, a gradient of GEM release from dendrimer

nanoparticles was generated by perfusing GEM-loaded nanoparticles (0.5 mg/mL) in one inlet,

and endothelial cell culture medium in the other inlet, using exactly the same experimental

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

201

source-sink mechanism used for the establishment and characterization of the nanoparticles

gradient. This was achieved by connecting the chip to the tubings into lateral inlets fed by glass

syringes (SGE), mounted on a Nexus 3000 syringe pump.

VII-3.10. Live/dead assay

The viability of cells cultured in the microfluidic chips in the presence or absence of GEM

loaded dendrimer nanoparticles was assessed using the Live/dead assay. For this, at endpoints,

the chip side-channels were washed with PBS (Sigma), and then incubated for 40 minutes with

Calcein AM (1 µM, live cells in green) and Ethidium homodimer 1 (6 µM, dead cells in red) and

then washed again. Microfluidic chips were then imaged with fluorescence microscope using

automated acquisition for Z-stack and multicolor-channel.

VII-3.11. Image processing

Acquired multi-frame multicolour images were combined either using Nikon AR or FIJI

software (open-source software ImageJ focused on biological image analysis) for operations of

processing such as rotation, crop, color balance and related assemblies, considering initial scales

and aspect ratios. High-resolution Z-stack images were processed for 3D reconstruction through

deconvolution and 3D projections made by NIS-Elements AR (Nikon) software.

VII-3.12. RT-PCR analysis

After each culture time, Matrigel from the central chamber was retrieved by punching with a

5 mm sterile biopsy punch (WPI) and placed in 24 well plates. Samples were washed with PBS

solution and Matrigel was depolymerized by incubation with 1 mL of MatriSperse Cell Recovery

Solution (BD Biosciences) for 30 minutes at 4 ºC. After complete release from the Matrigel, cells

were transferred to an Eppendorf tube and centrifuged to a pellet by centrifugation at 300 G for 5

minutes. Cells were washed with cold PBS and centrifuged again. Total RNA was extracted from

cells recovered from the 3D cultures using Phasemaker Tubes (Invitrogen), following the

manufacturer’s instructions. The pellet was then immersed in 500 mL of TRIZOL Reagent and

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

202

stored at -80 ºC until further use. RNA quantification and purity were assessed using a NanoDrop

ND-1000 spectrophotometer (NanoDrop Technologies). cDNA was synthetized according to the

protocol from iScript cDNA Synthesis Kit (Bio-Rad). The obtained cDNA was used as template for

the amplification of the target genes using the iQSYBR Green Supermix (Bio-Rad), according to

manufacturer’s instructions. The primers sequence, specific for each gene, namely Ribossomal

Protein S27a (RPS27A), Caspase-3 (Casp3), Ki-67 and Matrix metalloproteinase-1 (MMP-1) are

described in Table VII-1. The transcripts expression data were normalized to the endogenous

housekeeping gene RPS27A and the relative quantification was calculated normalizing the

obtained results for each target gene at each timepoint in basal culture conditions (controls),

according to the Livak (2ΔΔCT) method [24]. Three samples of each group were tested at each

time-point, in three independent experiments.

Table VIVII-1 – Primers list for the studied genes.

Oligo Name Sequence 5' to 3'

Casp-3 (human) F GAAATTGTGGAATTGATGCGTGA

Casp-3 (human) R CTACAACGATCCCCTCTGAAAAA

MMP-1 (human) F GGGGCTTTGATGTACCCTAGC

MMP-1 (human) R TGTCACACGCTTTTGGGGTTT

Ki67 (human) F GCTGGCTCCTGTTCACGTA

Ki67 (human) R CTGGGCTACACTGAGCACC

RPS27A (human) F GCTTGCCAGCAAAGATCAGT

RPS27A (human) R GAGGTTGAACCCTCGGATAC

VII-3.13. Immunohistochemistry

Microfluidic chip was processed for immunohistochemistry by perfusion of reagents, through

both side inlets. After cell culture, chip was washed with PBS, the cells were fixed with 4 %

paraformaldehyde in PBS for 30 min at room temperature (RT), and washed again with PBS.

Permeabilization of cells was done with 0.1 % (v/v) Triton X-100 (Sigma-Aldrich) for 15 minutes

at RT, followed by washing and incubation with blocking solution CAS-Block (Thermo Fisher

Scientific) for 30 minutes at 37 °C. Primary antibody for Ki-67 diluted in CAS-Block (rabbit,

ab833 R&D Systems, 1:100) was subsequently incubated overnight at 4 °C. Samples were

washed with PBS and incubated with the secondary antibody Alexa Fluor ® 488 conjugated

(1:500, anti-rabbit, R&D Systems) together with high-affinity filamentous actin probe phalloidin

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

203

conjugated with Alexa Fluor ® 594 (Invitrogen™, 1:80) in CAS-block for 3 hour at RT. The

samples were then washed with PBS, incubated with 4’,6-diamidino-2-phenylindole (Invitrogen™,

DAPI, 1:100 in PBS) for 10 minutes at RT, and finally washed twice with PBS. The samples were

kept in PBS in the dark at 4 °C until image analysis. High magnification images were acquired

with Nikon Eclipse Ti-E, using automated acquisition for Z-stack and multicolor-channel.

VII-3.14. Statistical analysis

Statistical analysis was performed using GraphPad Prism 5.0a software. The non-parametric

Mann–Whitney test was used to compare two groups. A value of p < 0.05 was considered

statistically significant. For multiple comparisons, the parametric tests One-Way ANOVA was

used, with p < 0.05 considered statistically significant.

VII-4. RESULTS

VII-4.1. Microfluidic chip fabrication and characterization

The aim of this work was to develop an in-house customized 3D microfluidic model that

emulates the human colorectal tumor microenvironment and enables the reconstitution of

physiological functions of microvascular tissue. The criteria for designing the model had to

comprise features that permitted the generation of soluble compounds’ gradient for the

assessment of the delivery efficiency of anti-cancer drug loaded nanoparticles. Moreover, easy-to-

use, possibility of real time imaging, performance of immunocytochemistry and cell retrieval after

the experiments for Real Time – Polymerase Chain Reaction (RT-PCR) analysis were

requirements of this platform. In addition to having to meet certain conditions in terms of

operation and design, the choice of material for fabrication, the geometry, dimensions of the

culture chamber and fluid flow control are some important considerations when designing the

device. The developed microfluidic chip is made using poly(dimethylsiloxane) (PDMS). The 3D

map of the final design is created by laser-reading the wafer’s surface at 50 x magnification, as

can be observed in Figure VII-1A. In this sense, our microfluidic device has three compartments:

a circular central chamber for the ECM like hydrogel (Matrigel) with 5 mm diameter and 126 µm

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

204

in depth, with a separate inlet and outlet. The central chamber is laterally sided by a pair of

perfused channels, 100 µm wide and 126 µm in depth, that are not interconnected between

them (Figure VII–1B).

To evaluate the maintenance of the co-culture over the course of 5 days, Matrigel supported

CRC cells were labeled with CellTrace™ Red (red) and HCoMECs with CellTrace™ Yellow (yellow),

as shown in Figure VII-1C). The endothelial cells were cultured in a 3D vessel-like fashion,

attaching to the walls of the lateral channels of the microfluidic chip. As expected, these cells

respond to a signal of VEGF that has been loaded in the Matrigel, by migrating from the lateral

channels towards the hydrogel through the interspacings between the pillars that separate the

two compartments. The “microvasculature” component was characterized by obtaining a 3D

representation (through actin/nuclei fluorescent stain of HCoMECs), to observe the morphology

of the colonic endothelial cells lining the lateral channel structure after 5 days under perfusion of

media (Figure VII–1D).

Figure VII-1 - Design and characterization of the microfluidic chip. A) Profilometer characterization: 3D map of the

microfluidic chip. Representative image of quality control and characterization of the microfluidic chip

features. The central chamber is 5 mm in diameter and 126 µm in depth, with a separate inlet and

outlet for hydrogel injection. The two lateral channels, each 100 µm wide and 126 µm deep, are not

interconnected, opening up the possibility of perfusing two distinct solutions. B) Schematic of chip

design and zoom in of the concept of colorectal tumor-on-a-chip model: round microfluidic central

chamber in which HCT-116 cancer cells are embedded in Matrigel; HCoMECs are seeded in the side

channels to form a 3D vessel-like assembly. C) Establishment of a microvascular 3D

microenvironment of colorectal tumor-on-a-chip: HTC-116 CRC cells embedded in Matrigel (in central

chamber, stained in red) and HCoMECs (in lateral microchannels, stained in yellow) at day 1 and 5

of culture. Endothelial cells start to invade the central chamber filled with Matrigel in response to

VEGF presence. D) Fluorescence microscopy image of microfluidic lateral channel mimicking pre-

vascularization with HCoMECs cells after 5 days of culture (DAPI blue – Nuclei; Phalloidin green – F-

actin filaments); Representative fluorescence microscopy close-up image of microchannel cross

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

205

section showing endothelial cells aligning and creating an endothelialized lumen within the

microchannel.

VII-4.2. Endothelial Invasion

The three-dimensional formation of endothelial sprouts in a microfluidic device and the

changes in cellular organization during early stages of invasion were examined. This invasion is

oriented directly towards the source of VEGF in the Matrigel, despite the fact that cell migration

from the endothelium is not physically constrained in any direction by the system design (Figure

VII–2A). Sprouting quantification shows that HCoMECs could progress up to 500 µm after 5

days. The leading tip cells is replete with filopodia-like protrusions, morphologically recapitulating

in vivo endothelial-cell sprouting. As the sprouts continued to invade and extend throughout the

3D matrix, they become longer, contain progressively more cells, and begin to form

interconnected branches, as it occurs in early stages of tumor development [25] (Figure VII-2B,

zoom in at day 3). Although we are not mimicking the angiogenesis phenomena per se on this

model, we observe that endothelial cells invade the hydrogel core but do not form anastomosis.

Figure VII-2 - Endothelial cell invasion. A) Schematic representation of HCoMECS cells in the lateral channels

invading the central chamber in response to the presence of VEGF mixed in the Matrigel; B)

Formation of endothelial sprouts in the microfluidic device: representative bright field images of the

chip taken at determined time points were analyzed in Image J. C) Quantification of endothelial

invasion. Data is presented as mean ± stdev (n=9), (*) denotes statistical differences (p < 0.01).

VII-4.3. Validation of the integrated microfluidic gradient chip device

Ultimately, in this work, we aim to establish a 3D cancer microfluidic cell culture for onco-

nanomedicine efficacy. Through its architecture, the platform offers the possibility to observe the

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

206

effect of several concentrations of GEM-loaded CMCht/PAMAM dendrimer nanoparticles, as they

penetrate the HCT-116 laden-hydrogel in a gradient fashion. As a result, from one experiment, a

real-time response of the cancer cells in a gradient-like fashion is obtained, as occurs in a

physiological concentric tumor microenvironment. For this purpose, the formation of the gradient

was validated using fluorescent labeled CMCht/PAMAM dendrimer nanoparticles prior to the

GEM release. It is noteworthy that these nanoparticles present remarkable physicochemical and

biological properties such as solubility in water, biocompatibility and ability to be easily surface

modified. Therefore they have great potential to be used in fundamental cell biology studies as

well as for intracellular drug delivery [26].

The CMCht/PAMAM dendrimer nanoparticles were synthetized and characterized previously,

showing an average size of 50 nm [22, 27] (Figure VII–3A). As depicted in Figure VII–3B), after

60 minutes, a gradient of fluorescence was established in the cell culture chamber, and

maintained for at least 12 hours of continuous lateral perfusion, exhibiting a quasi-linear

fluorescent profile (from 100 % at the source to 0 % at the sink) (Figure VII-3C). A 3D projection of

FITC-labeled nanoparticles’ gradient according to surface fluorescence intensity (Figure VII–3D)

also confirms these findings.

Figure VII-3 - Schematic representation of fluorescent nanoparticle gradient formation. A) Transmission electron

microscopy image of individual CMCht/PAMAM dendrimer nanoparticle, representative image of chip

connected to tubing perfusing the nanoparticle solution, and zoom in confocal microscopy image

confirming FITC-labeled CMCht/PAMAM dendrimer nanoparticles’ dispersion in Matrigel. B) Panel of

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

207

fluorescent microscope images of FITC-labeled nanoparticles’ gradient up to 12 hours. C)

Quantification of fluorescence: intensity (arbitrary units) vs distance (µm) across the chip up to 12

hours (measurements according to the direction of the green arrow indicated in B). D) 3D projection

of FITC-labeled nanoparticles’ gradient according to fluorescence intensity.

VII-4.4. Drug Screening Validation

We aimed to establish an image-based method to investigate the concentration-dependent

response of cells inside the microfluidic device by creating a gradient of concentrations of GEM.

To investigate the concentration-dependent toxicity on cancer cells inside the microfluidic chip,

the survival response of cells of the CRC cell line HCT-116 embedded within Matrigel was

examined. The cells were allowed to adhere and develop for 12 hours prior to drug exposure. The

exposure of the cells to the delivery-induced gradient of GEM was generated from a maximum

concentration of 0.5 mg/mL of dendrimer nanoparticles sourced from one microvasculature

lumen perfused with them and diffused through the hydrogel core towards the opposite side

(Figure VII-3).

GEM-loaded nanoparticles were studied in terms of encapsulation efficiency and drug-

loading efficiency. The encapsulation efficiency of GEM-CMCht/PAMAM was 52.99 % ± 3.5 %,

and the drug-loading efficiency was 9.42 % ± 3.5 %. The in vitro release profiles of GEM-loaded

CMCht/PAMAM were recorded in a cumulative release curve during 5 days. As shown in Figure

VII-4, after 6 hours, around 40 % of the drug was released, which shows that GEM-loaded

CMCht/PAMAM can retain their stability in solution. Thereafter, the nanoparticles showed an

accelerated release, where more than half of the GEM was released by 24 hours (Figure VII-4A).

In addition, an important event to notice is the immediate internalization of the nanoparticles

after contact with cells, as shown in our previous study [27], decreasing the chance of drug

degradation. Such intracellular release mechanism represents a key factor of the significance of

GEM loaded nanoparticles as an anti-tumoral agent [28].

Live experiments were similarly performed using the same type and concentration profiles

for the sake of reproducibility and final measurement co-relation. In order to quantify cells’

response to GEM, we segmented the circular area of the chip (core) into 6 concentric areas, as

represented in Figure VII-4B. Thereby, the following nomenclature was used: M1, M2, M3 was

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

208

assigned to the areas of the chip subjected to culture medium (Figure VII-4B, left side), and D1,

D2 and D3 to the areas of the chip subjected to drug (Figure VII-4B, right side).

After 1 and 5 days of exposure to GEM, the viability of CRC cells was evaluated in the 3D

matrix, under perfusion at side-channels. The controls, at both time points (Figure VII-4C/D)

indicate insignificant levels of cell death, demonstrating the microfluidics’ model ability to support

a stable and viable co-culture of HCoMECs and HCT-116 CRC cells in contiguous compartments

under perfusion. Cell viability in controls show that oxygen can freely permeate to the cells either

through PDMS or the hydrogel matrix, while being fed for nutrients dynamically by perfusion of

side channels (Figure VII - 4B). On the other hand, in the presence of GEM-loaded

CMCht/PAMAM dendrimer nanoparticles, cell death occurs in a gradient manner, with a step-

wise decay in toxicity, i.e. less linear than the concentration release profile of GEM previously

quantified. This indicates that GEM could be efficiently released from the nanoparticles and taken

up by the cancer cells while remaining an active anti-cancer drug, in its soluble form (Figure VII-

4C). The cell death induced by the anti-cancer drug was more efficient within the first 150 µm of

the cell culture, which might be due to an eventual threshold of drug efficacy on cells. This

concentration effect has been consistent over time, with the cell death values increasing from day

1 to day 5 (Figure VII-4C).

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

209

Figure VII-4 - GEM drug effect on colorectal-tumor-model. A) Release profile of GEM: cumulative release (hours) of

GEM at pH 7.4 in PBS, at 37 ºC and stirred at 60 rpm, determined by UV spectrophotometer set at

275 nm. The results are expressed as mean ± stdev, n=6. B) Schematic representation of

experimental set-up (perfusion of culture medium through one in- and one outlet and of culture

medium supplemented with 0.5 mg/mL GEM-loaded dendrimer nanoparticles through the other in-

and outlet); the areas defined for cell death quantification are delimited by dashed lines and each

ring assigned D1-D3 and M1-M3. C) Representative panels of fluorescence microscopic images of

live/dead assay performed on microfluidic chip at days 1 and 5. D) Quantification of cell death,

represented as percentage, based on the microscopy data. Data is presented as mean ± stdev (n=3),

(*) denotes statistical differences (p < 0.05). Scale bar 1000 µm.

VII-4.5. Gene Expression Analysis and Immunocytochemistry on a Chip

To the best of our knowledge, very few organ-on-chip models allow for gene expression

analysis of the cells cultured therein due to the small mRNA content accessible from the chip’s

culture chambers. On a sub-set of corresponding experiments, exposure to GEM-loaded

CMCht/PAMAM dendrimer nanoparticles was tested as single-concentration exposure per device

(GEM-loaded CMCht/PAMAM perfused through both inlets), not as a gradient, to demonstrate the

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

210

potential of the platform for evaluating gene expression as function of dose-response. To this end,

two representative conditions of previous experiments, i.e. 0 and 0.5 mg/mL (on continuously

perfused media) were tested as proof-of-concept for the suitability of the model for RT-PCR and

gene studying. 0 mg/mL represents our control group. At each time point (1 and 5 days), cells

were extracted from the device, and RNA was extracted from cells – after hydrogel removal

(Figure VII-5A) for analysis of mRNA expression of Caspase-3, Ki-67 and MMP-1 by real time

PCR. These markers where chosen to allow the evaluation of the effect of cytotoxic compounds:

MMP-1 because of its relationship with the metastatic phenotype [29], Caspase-3 because it is

involved in apoptosis, and Ki-67 as a universal marker for proliferation.

In general, a down-regulation is observed for all genes when comparing day 1 to day 5, as

shown in Figure VII-5. Compared to the controls, MMP-1 expression is down-regulated, 0.5- and

0.2-fold for day 1 and 5, respectively. Similarly, for Caspase-3, at day 1, no statistical differences

between the control and GEM-treated cells is observed. For Ki-67 expression, no significant effect

of the exposure to GEM was observed at day 1, which can be due to the limited time for the

compound to act on the cells and affect their proliferation. However, at day 5 a statistical

significant effect of GEM is observed, with a 0.4 fold decrease in the expression of Ki-67

compared to the control. These results demonstrate the suitability of our platform for assessing

gene expression.

Concerning Ki-67 immunofluorescence on the microfluidic chip shown in Figure VII-5C), the

marker is expressed only when the nanoparticles are absent, which confirms the PCR data.

These qualitative results translate the presence of Ki-67-positive cells, corresponding to

proliferating tumor cells in the area of the chip more subjected to drug-free media, in contrast to

Ki-67-negative cells present in the matrix regions close to the perfusion source of GEM loaded

CMCht/PAMAM dendrimer nanoparticle media.

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

211

Figure VII-5 - Gene expression analysis and immunocytochemistry on a chip. A) Representation of hydrogel retrieval

containing HCT-116 cells from the chip for gene expression analysis. B) Gene expression of Ki-67,

Caspase-3 and MMP-1, in the absence (control) and presence of gemcitabine, at day 1 and day 5.

The values for controls are represented by the dashed line. * Significant differences when comparing

the same condition at two different time points; # significant differences when comparing to controls

at the same time point. Data represent mean values of 3 independent experiments ± stdev. C)

Immunohistochemistry: in the center, a representative image of the entire chip is depicted. The

lateral images represent the zoom in at the cell level showing the immunocytochemistry Ki-67

staining after 5 days in culture (panels showing DAPI nuclei staining blue, Ki-67 staining green, F-

actin staining red and merged image in colors).

VII-5. DISCUSSION

The design of our new microfluidic chip resembles the traditional ECM-containing devices,

where endothelial monolayers are cultured onto the sidewall of an extracellular matrix material,

that separates two microfluidic channels [30]. In our model, the “tumor” core defined by the

hydrogel is supported by an adjacent “microvascular” network fed by external syringe pumps,

supporting physiological events specific of each cells type. The lumen-like channels deliver

nutrients to surrounding tissues sufficient to support robust growth. This opens up the possibility

of perfusing two distinct solutions in each side of the central chamber containing Matrigel and the

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

212

cancer cells encapsulated therein, and therefore the ability to produce and maintain stable

gradients. Around the inner part of the chamber we designed micro pillars to help retain Matrigel

when injected into the chamber, creating enough superficial tension and avoiding leakage

towards the lateral channels (Figure VII-1C). Furthermore, the liquid continuity between the side

channels and the culture chamber was ensured by these small pillars, which act as resistors,

preventing the shear stress to be transmitted to the cell culture chamber [31]. The chip is made

using PDMS due to its several unique properties, which make it a suitable choice for microfluidic

application: PDMS is inexpensive, easy to handle, gas permeable (assuring the supply of oxygen

to the cells), and transparent, properties needed for the imaging [32]. Corroborating this, our

results show the CRC cells in red and endothelial cells in yellow clearly proliferating, elongating in

the direction of the flow and covering the whole lateral channels, consequently forming a vessel-

like structure at day 5. Interestingly, we observed an endothelial 3D macro structure invading the

central chamber through the pillars, in response to the VEFG, defining an early stage of sprouting

(Figures VII-1C and D).

When characterizing the microfluidic co-culture, the sprouting phenomena caught our

attention. Angiogenesis is a complex morphogenetic process where endothelial cells from pre-

existing vessels invade the matrix as multicellular sprouts to form new vessels [33]. For capillary

sprouting and blood vessel formation, both proliferation and migration are important, but how

these two events are coordinated during the extension process is not fully clear [34]. Pro-

angiogenic signals, such as VEGF, promote processes such as proliferation, endothelial motility

and filopodia extension [35]. Evidence from preclinical and clinical studies indicates VEGF is the

predominant angiogenic factor in human colon cancer and is associated with formation of

metastases and poor prognosis [36]. So far, innumerous models tried to recreate this

phenomena [34, 37, 38]. For instance, the tube formation assay involves the reorganization of

endothelial cells seeded at Matrigel’s surface to form cords similar to vascular networks. Still they

lack important features observed in native angiogenesis including directional invasion of cells into

a 3D extracellular matrix (ECM), proper polarization of the luminal, lumen formation, and support

of fluid flow [39]. Although it has been stated before that we are not trying to recreate an

anastomosis in the central chamber, the fact that we cannot observe it can be explained by

several factors, for example, by the fact that the central chamber of the device is too large to

observe such events. Nevertheless, from a technical perspective, this feature allows higher

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

213

volumetric hosting of HCT-116 cells, which enables quantitative analysis of gene expression on

this tissue (i.e., enough cells for PCR analysis). Another factor is probably the lack of support

cells, such as perycites. This cell type would help to stabilize vessel formation and therefore

permit the bridging of sprouts over longer distances of cell culture regions [40]. As our goal was

uniquely to use the side channels to create a perfusable lateral channel with a lumen, we seeded

the endothelial cells in the lateral channel, and not in the central chamber. However, it is

interesting to see the natural behavior of the endothelial cells invading the Matrigel filled chamber

with the cancer cells and further characterize our model. As conclusion, this biomimetic model is

able to reconstitute the beginning of the angiogenic sprouting process.

To generate a stable gradient, one of the main goals of this project, the CMCht/PAMAM

dendrimer nanoparticles were chosen. Their proven ability to act as drug release vehicles inside

the cells, as well as cell trackers (when surface modified with a fluorescence molecule such as

FITC [22, 27]) made them excellent candidates. One of the initial concerns was the ability of

these nanoparticles to perfuse through Matrigel’s fiber network. As observed in Figure VII–3A, the

nanoparticles disperse widely when perfused in the lateral channels, in a gradient fashion. The

real time observation and establishment of a fluorescent gradient generated by the perfusion of

the FITC-labeled nanoparticles was essential to move on to drug screening assays, using the

same flow and the same nanoparticles, this time releasing the drug overtime.

Regarding the GEM release from the dendrimer nanoparticles in the microfluidic model, a

concentration of 0.5 mg/mL was chosen with the rationale to a) avoid the cytotoxic effect of the

nanoparticles and b) enable continuous and sustained intracellular drug delivery by the

nanoparticles and therewith a high efficacy, as was demonstrated by the viability results (Figure

VII-4) and qPCR data (Figure VII- 5). Also, nanoparticles are highly advantageous when comparing

to the golden standard clinical delivery of one bulk delivery (every two weeks), more prone to

result in secondary undesirable effects.

Given that the gradient we observed was more or less concentrically generated from the

periphery towards the center of the chamber, the areas defined for quantification followed this

geometry. Apparently, lower amounts of nanoparticles leading to drug release in the areas further

away from the source are not sufficient to induce cell death. Higher drug efficacy at D1-D2

regions seems to define a cut-off area on the action range of the nanomedicines in the matrix

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

214

delineated by the Matrigel cell-culture. Another important aspect to consider when analyzing the

data is that although we were not quantifying cell death among endothelial cells, it is visible that

both the cancer cells and the vasculature are sensitive to standard-of-care chemotherapeutic

Gemcitabine. Targeting of the nanoparticles towards CRC cells and minimizing deleterious effects

on endothelial cells would be of extreme usefulness and constitutes the next step of this work. No

significant differences are observed between the controls at both days or between the controls

and the areas subjected to culture medium M1, M2 and M3. It is interesting to note that even

after 5 days of exposure to the maximum concentration of drug possible in this model (area D1),

only about 40 % of cell death was observed. This low cell death level is in agreement with

previous studies, corroborating the fact that 3D cultures usually show a reduced sensitivity to

anticancer drugs [41]. This can be due to various factors, including matrix stiffness, which was

reported to be closely related to tumor chemoresistance decreased penetration of anti-cancer

drugs, increased pro-survival signaling, and/or up-regulation of genes conferring drug resistance.

Moreover, cells could be in the process of dying at the analysis time, but not yet dead and

therefore not quantified in this screening, since the ethidium homodimer only stains dead cells.

This microfluidic chip also allows the use of other stainings.

To the best of our knowledge, very few organ-on-chip models allow for gene expression

analysis of the cells cultured therein due to the small mRNA content accessible from the chip’s

culture chambers. Moreover, the need to use hydrogels such as Matrigel to replicate the intricate

3D microenvironment in which cells live makes it harder to retrieve them for genetic analysis in

good conditions. This analysis works as a proof-of-concept to demonstrate the versatility of this

platform to work both as a drug screening platform and allowing genetic analysis. One of the

main challenges of this task was to effectively retrieve HTC-116 cells from the Matrigel inside the

5 mm central chamber of the microfluidic device. The strategy we developed involves punching

out the central chamber with a sterile surgical 5 mm punch and dissolving the Matrigel with

encapsulated cancer cells in Recovery Solution. Results show a down regulation of all genes

tested. MMP-1 is a collagenase that degrades ECM, specifically targeting type I, II and III

collagens, the major components of the interstitial stroma [42]. It has been suggested to be

associated with an advanced stage of colon cancer with poor prognosis and has emerged as a

new oncogenesis and metastasis target for treatment of diverse cancers [43]. However,

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

215

molecular factors that contribute directly to tumor cell vascular penetration and the extent of

MMP-1 activity in the presence of chemotherapeutic drugs has not been fully identified [44, 45].

For example, Khanna et al. [46] used small interfering RNA (siRNA) against proto-oncoprotein

Ets-1 in gemcitabine-resistant cells, and demonstrated a reversal in gemcitabine

chemosensitivity, translated in a marked reduction in the expression of MMP-1[46]. Given this

evidence, it makes sense that exposing HCT-116 cells to an anti-cancer drug would lead to the

down regulation of the expression of MMP-1, as observed, however, with no statistical

differences. Regarding caspase-3 results, the down regulation obtained over the two timepoints

corroborates the previous live/dead assay results, where at day 1 the extent of cell death due to

GEM exposure is less pronounced than day 5 (although with no statistical differences when

compared to the respective control). Nevertheless, at the later time point, the expression of

Caspase-3 is downregulated, even though we see a clear increase in cell death in the live/dead

experiments performed before. Recently, caspase activity identified in the absence of cell death

has sparked strong interest in caspase functions in non-apoptotic cellular responses, suggesting

that caspases can be activated without inducing or even before apoptosis occurs [47]. Several

pathways can be related to GEM-induced apoptosis. On one hand, literature evidence exists for

caspase-dependent apoptosis [48], while other studies describe GEM-induced apoptosis via

reactive oxygen species generation, without increase in Caspase-3 [49]. In addition, GEM-induced

apoptosis is also related to signal-regulated kinase (ERK), Akt, Bcl-2 and p38 mitogen-activated

protein kinase (MAPK) pathways, and not only Caspase-3 [50].

When looking at Ki-67 expression, a significant difference in cells exposed to the drug from

day 1 to day 5 reveals that the drug decreased cell proliferation. Its increased expression in

human cancer specimens generally denotes an aggressive phenotype, although new studies

suggest that this process is more complex than initially thought [51]. However, the fact that many

aggressive cancers have high expression of Ki-67 relative to normal and quiescent cells suggests

that Ki-67 might be an attractive target for cancer therapy if its role as a driver of carcinogenesis

were validated.

The developed tumor-on-chip platform comprising a human CRC-like core and surrounding

vascularized microtissue is a promising tool to be used in onco-medicine, either for high-content

image-based screenings or gene expression analysis to study drug-dose responses. Based on a

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

216

robust dynamic microfluidic system with a viable co-culture of HTC-116 CRC cells and

HCoMECS, this platform was validated as tumor-on-chip in vitro platform suitable for assessing

an anti-cancer nanoparticle-based deliverable.

Although very useful as a high-throughput system for drug testing, our model does not

comprise an oxygen gradient, common to the in vivo cancer microenvironment. Generating such

a gradient would require inclusion of another complex feature in this already complex model, and

will be considered as a part of the future work. In this study, the in-house-synthetized dendrimer

nanoparticles were used as a model for nanomaterial based anti-cancer therapeutics. The

fluorescently-modified nanoparticles successfully acted as a tracking tool that allowed the real-

time imaging of the nanoparticle location, whereas drug-loaded particles functioned as an

effective delivery vehicle of GEM in a gradient fashion. The resulting data sets enabled the

definition of precise ranges of action of these anti-cancer nanoparticles, taking into account the

complexity of the tumor microenvironment, including the critical microvasculature component.

Highlights the importance of the high range of concentrations in a single experiment at this

model, which has an impact on its robustness as a validation tool This tumor-on-chip device

enabled both the performance of RT-PCR and immunocytochemistry stainings as the drug was

being delivered to the tumor core. Technically, we proved that cells can easily be extracted from

the chip and studied in terms of gene expression, providing valuable information on the

mechanisms underlying the drug effect. Through this 3D microfluidic cell culture, the study of

phenomena such as vascularization and oncogenesis in dynamic conditions may provide a

powerful insight on the cancer stage prognosis and eventually on depicting a treatment option.

Such capabilities are of high value, especially if using patient cells, and, thus, for precision

therapy screenings on personalized medicine.

Most of all, the greatest advantages of our platform compared to others of the same type

(central chamber for 3D matrix laterally sided by perfusable channels) are: 1) facile molecular

cell analysis; 2) physiologically inspired design of radial drug penetration into solid tumors; 3)

hybrid co-culture system to show efficiency of drug penetration through a micro-vascular network

into a cancer-mimicking tissue, comprised of a supporting relevant matrix with colorectal cancer

cells; 4) real-time imaging of drug delivery and its effect on cell viability; 5) design compatible

with assessing physiologically-relevant cell migration, thus metastatic events (not the scope of the

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

217

current manuscript); and 6) system easily combined with organoid seeding of the central

chamber.

VII-6. CONCLUSIONS

This platform has the potential to further study the EPR effect by controlling the permeability

of the vascular network, e.g. by readjusting the distance between pillars. Moreover, tumor cell

stress can be studied by e.g. varying the flow rates and, thus, exposing the cells to shear

stresses. Also, while tumor cell signaling was not studied here, this could be done by analyzing

the medium collected in the outlet for e.g. soluble factors profiling. The potential for further

application is vast, from testing various types and sizes of particles and patient-derived cells, to

bringing this tool closer to a clinical and pharmaceutical translatable setting in further research.

Taken together, this platform can potentially lead to breakthrough discoveries not only in

basic research but also related to clinical applications, namely in CRC field.

VII-7. ACKNOWLEDGMENTS

The project FROnTHERA (NORTE-01-0145-FEDER-000023), supported by Norte Portugal

Regional Operational Programme (NORTE 2020), under the PORTUGAL 2020 Partnership

Agreement, through the European Regional Development Fund (ERDF). Mariana Carvalho

acknowledges her PhD scholarship NORTE-08-5369-FSE- 000044, funded by Programa

Operacional Regional do Norte, Fundo Social Europeu, Norte2020 TERM&SC and EMBO Short-

Term Fellowship 7232. J. M. Oliveira thanks FCT for his distinction attributed under the FCT

Investigator program (IF/00423/2012 and IF/01285/2015. This work was partially supported

by the IET Harvey Engineering Research Award 2018 (ENG ThE CANCER) and with the support of

the Dutch Province of Limburg.

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

218

VII-8. REFERENCES

1. Vatandoust, S., T.J. Price, and C.S. Karapetis, Colorectal cancer: Metastases to a single organ. World Journal of Gastroenterology, 2015. 21(41): p. 11767-11776. 2. Carvalho, M.R., et al., Anti-Cancer Drug Validation: the Contribution of Tissue Engineered Models. Stem Cell Rev, 2017. 3. Phan, D.T.T., et al., A vascularized and perfused organ-on-a-chip platform for large-scale drug screening applications. Lab Chip, 2017. 17(3): p. 511-520. 4. Barata, D., C. van Blitterswijk, and P. Habibovic, High-throughput screening approaches and combinatorial development of biomaterials using microfluidics. Acta Biomaterialia, 2016. 34(Supplement C): p. 1-20. 5. Foudeh, A.M., et al., Microfluidic designs and techniques using lab-on-a-chip devices for pathogen detection for point-of-care diagnostics. Lab Chip, 2012. 12(18): p. 3249-66. 6. Orcheston-Findlay, L., et al., A microfluidic gradient generator to simulate the oxygen microenvironment in cancer cell culture. Microelectronic Engineering, 2018. 195: p. 107-113. 7. Bott, K., et al., The effect of matrix characteristics on fibroblast proliferation in 3D gels. Biomaterials, 2010. 31(32): p. 8454-64. 8. Barata, D., C. van Blitterswijk, and P. Habibovic, High-throughput screening approaches and combinatorial development of biomaterials using microfluidics. Acta Biomaterialia, 2016. 34: p. 1-20. 9. Ma, Y.-H.V., et al., A review of microfluidic approaches for investigating cancer extravasation during metastasis. Microsystems &Amp; Nanoengineering, 2018. 4: p. 17104. 10. Wan, L., et al., Mimicking Embedded Vasculature Structure for 3D Cancer on a Chip Approaches through Micromilling. Scientific Reports, 2017. 7(1): p. 16724. 11. Chen, M.B., et al., On-chip human microvasculature assay for visualization and quantification of tumor cell extravasation dynamics. Nat Protoc, 2017. 12(5): p. 865-880. 12. Zervantonakis, I.K., et al., Three-dimensional microfluidic model for tumor cell intravasation and endothelial barrier function. Proc Natl Acad Sci U S A, 2012. 109(34): p. 13515-20. 13. Berthier, E. and D.J. Beebe, Gradient generation platforms: new directions for an established microfluidic technology. Lab Chip, 2014. 14(17): p. 3241-7. 14. Keenan, T.M. and A. Folch, Biomolecular gradients in cell culture systems. Lab on a chip, 2008. 8(1): p. 10.1039/b711887b. 15. Schwarz, J., et al., A microfluidic device for measuring cell migration towards substrate-bound and soluble chemokine gradients. Scientific Reports, 2016. 6: p. 36440. 16. Fouad, Y.A. and C. Aanei, Revisiting the hallmarks of cancer. Am J Cancer Res, 2017. 7(5): p. 1016-1036. 17. Shangguan, W., et al., Endothelium originated from colorectal cancer stem cells constitute cancer blood vessels. Cancer Science, 2017. 108(7): p. 1357-1367. 18. Jiménez-Fonseca, P., et al., Gemcitabine plus capecitabine (Gem–Cape) biweekly in chemorefractory metastatic colorectal cancer. Clinical & Translational Oncology, 2015. 17(5): p. 384-392. 19. ClinicalTrials.gov. Gemcitabine in Treating Patients With Advanced Colorectal Cancer. 2017 23th May 2018]; Available from: https://clinicaltrials.gov/ct2/show/NCT00007943.

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

219

20. Yu, X., et al., An in vitro and in vivo study of gemcitabine-loaded albumin nanoparticles in a pancreatic cancer cell line. International Journal of Nanomedicine, 2015. 10: p. 6825-6834. 21. Saif, M.W., et al., The efficacy of gemcitabine as salvage treatment in patients with refractory advanced colorectal cancer (CRC): a single institution experience. Anticancer Res, 2011. 31(9): p. 2971-4. 22. Oliveira, J.M., et al., Surface Engineered Carboxymethylchitosan/Poly(amidoamine) Dendrimer Nanoparticles for Intracellular Targeting. Advanced Functional Materials, 2008. 18(12): p. 1840-1853. 23. Chen, X.-G. and H.-J. Park, Chemical characteristics of O-carboxymethyl chitosans related to the preparation conditions. Carbohydrate Polymers, 2003. 53(4): p. 355-359. 24. Livak, K.J. and T.D. Schmittgen, Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2−ΔΔCT Method. Methods, 2001. 25(4): p. 402-408.

25. Jacquemet, G., H. Hamidi, and J. Ivaska, Filopodia in cell adhesion, 3D migration and cancer cell invasion. Current Opinion in Cell Biology, 2015. 36: p. 23-31. 26. Oliveira, J.M., et al., The osteogenic differentiation of rat bone marrow stromal cells cultured with dexamethasone-loaded carboxymethylchitosan/poly(amidoamine) dendrimer nanoparticles. Biomaterials, 2009. 30(5): p. 804-13. 27. Carvalho, M.R., et al., A semiautomated microfluidic platform for real-time investigation of nanoparticles' cellular uptake and cancer cells' tracking. Nanomedicine (Lond), 2017. 12(6): p. 581-596. 28. Han, H., et al., Enzyme-sensitive gemcitabine conjugated albumin nanoparticles as a versatile theranostic nanoplatform for pancreatic cancer treatment. Journal of Colloid and Interface Science, 2017. 507: p. 217-224. 29. Bendardaf, R., et al., MMP-1 (collagenase-1) expression in primary colorectal cancer and its metastases. Scand J Gastroenterol, 2007. 42(12): p. 1473-8. 30. Bogorad, M.I., et al., Review: in vitro microvessel models. Lab on a Chip, 2015. 15(22): p. 4242-4255. 31. Barata, D., et al., Development of a shear stress-free microfluidic gradient generator capable of quantitatively analyzing single-cell morphology. Biomed Microdevices, 2017. 19(4): p. 81. 32. Markov, D.A., et al., Variation in diffusion of gases through PDMS due to plasma surface treatment and storage conditions. Biomed Microdevices, 2014. 16(1): p. 91-6. 33. Nishida, N., et al., Angiogenesis in Cancer. Vascular Health and Risk Management, 2006. 2(3): p. 213-219. 34. Yeon, J.H., et al., In vitro formation and characterization of a perfusable three-dimensional tubular capillary network in microfluidic devices. Lab Chip, 2012. 12(16): p. 2815-22. 35. Weis, S., et al., Endothelial barrier disruption by VEGF-mediated Src activity potentiates tumor cell extravasation and metastasis. J Cell Biol, 2004. 167(2): p. 223-9. 36. Bendardaf, R., et al., The effect of vascular endothelial growth factor-1 expression on survival of advanced colorectal cancer patients. The Libyan Journal of Medicine, 2017. 12(1): p. 1290741. 37. Tahergorabi, Z. and M. Khazaei, A Review on Angiogenesis and Its Assays. Iranian Journal of Basic Medical Sciences, 2012. 15(6): p. 1110-1126. 38. Song, W., et al., Tumor-derived extracellular vesicles in angiogenesis. Biomedicine & Pharmacotherapy, 2018. 102: p. 1203-1208.

Chapter VII – Colorectal Tumor-on-a-chip System: A 3D Tool for Precision Onco-nanomedicine

220

39. Nguyen, D.H., et al., Biomimetic model to reconstitute angiogenic sprouting morphogenesis in vitro. Proc Natl Acad Sci U S A, 2013. 110(17): p. 6712-7. 40. Nguyen, D.-H.T., et al., Biomimetic model to reconstitute angiogenic sprouting morphogenesis in vitro. Proceedings of the National Academy of Sciences, 2013. 110(17): p. 6712-6717. 41. Wen, Z., et al., A spheroid-based 3-D culture model for pancreatic cancer drug testing, using the acid phosphatase assay. Braz J Med Biol Res, 2013. 46(7): p. 634-42. 42. Shen, C.J., et al., MMP1 expression is activated by Slug and enhances multi-drug resistance (MDR) in breast cancer. PLoS One, 2017. 12(3): p. e0174487. 43. Foley, C.J., et al., Matrix metalloprotease-1a promotes tumorigenesis and metastasis. J Biol Chem, 2012. 287(29): p. 24330-8. 44. Benbow, U., et al., Selective modulation of collagenase 1 gene expression by the chemotherapeutic agent doxorubicin. Clin Cancer Res, 1999. 5(1): p. 203-8. 45. Said, A.H., J.-P. Raufman, and G. Xie, The Role of Matrix Metalloproteinases in Colorectal Cancer. Cancers, 2014. 6(1): p. 366-375. 46. Khanna, A., et al., Ets-1 expression and gemcitabine chemoresistance in pancreatic cancer cells. Cell Mol Biol Lett, 2011. 16(1): p. 101-13. 47. Kuranaga, E., Beyond apoptosis: caspase regulatory mechanisms and functions in vivo. Genes Cells, 2012. 17(2): p. 83-97. 48. Chandler, N.M., J.J. Canete, and M.P. Callery, Caspase-3 drives apoptosis in pancreatic cancer cells after treatment with gemcitabine. J Gastrointest Surg, 2004. 8(8): p. 1072-8. 49. Guo, Y., et al., Overexpression of heat shock protein 27 (HSP27) increases gemcitabine sensitivity in pancreatic cancer cells through S-phase arrest and apoptosis. J Cell Mol Med, 2015. 19(2): p. 340-50. 50. Kao, Y.-T., et al., Involvement of p38 mitogen-activated protein kinase in acquired gemcitabine-resistant human urothelial carcinoma sublines. The Kaohsiung Journal of Medical Sciences, 2014. 30(7): p. 323-330. 51. Melling, N., et al., High Ki67 expression is an independent good prognostic marker in colorectal cancer. J Clin Pathol, 2016. 69(3): p. 209-14.

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

221

Chapter VIII

Peptide-modified Dendrimer

Nanoparticles for Targeted

Therapy of Colorectal Cancer

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

222

Chapter VIII

Chapter VIII - Peptide-modified Dendrimer Nanoparticles for

Targeted Therapy of Colorectal Cancer1

ABSTRACT

Peptides have recently emerged as a promising class of targeting ligands for specific drug delivery in

cancer treatment, which avoid undesirable side effects of the systemic administration of

chemotherapeutics. Their conjugation with nanoparticles have demonstrated to improve the

functionality of peptides resulting in a versatile platform for biomedical applications. In this work, we

describe the development of carboxymethylchitosan/poly(amidoamine) (CMCht/PAMAM) dendrimer

nanoparticles functionalized with YIGSR laminin receptor binding peptide for the active targeting and

specific delivery of therapeutic agents into colorectal cancer cells. The successful functionalization was

confirmed by several physico-chemical characterization techniques. The selectivity of the YIGSR-

CMCht/PAMAM dendrimer nanoparticles was first validated in vitro using a micropatterned array of

67kDa laminin receptor. Next, the specificity of YIGSR-CMCht/PAMAM dendrimers nanoparticles

towards laminin receptor was further confirmed both in 2D and 3D settings using HCT-116 colorectal

cancer cells and L929 fibroblasts in co-culture. Finally, gemcitabine-loaded YIGSR-CMCht/PAMAM

dendrimer nanoparticles induced a targeted mortality on HCT-116 cancer cells in a co-culture scenario.

Overall, our study shows solid evidences that YIGSR laminin receptor binding peptide coupled to

CMCht/PAMAM dendrimer nanoparticles may be employed as an anti-cancerous target for the specific

and intracellular delivery of chemotherapeutic agents.

Keywords: Targeting; Dendrimer nanoparticles; Colorectal cancer; Micropatterning.

1This chapter is based on the following publication:

M. R. Carvalho, C. R. Carvalho, F.R. Maia, D. Caballero, S. C. Kundu, R. L. Reis, J. M. Oliveira. “Peptide-modified dendrimer nanoparticles

for targeted therapy of colorectal cancer” Advanced Therapeutics, 2019, 1900231. doi: 10.1002/adtp.201900132.

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

223

VIII-1. GRAPHICAL ABSTRACT

VIII-2. INTRODUCTION

Cancer is a leading cause of death worldwide, with more than 10 million new cases each year, and

an estimation of ~13.1 million cancer-related deaths by 2030 [1]. Typically, cytotoxic chemotherapies

have a narrow therapeutic window, with high peaks of plasma concentration. This leads to high-dose

requirements with adverse side effects, reduced therapeutic indices, non-specific targeting, and

development of multiple drug resistance [2].

Nanotechnology has evolved as an exciting platform in the field of cancer research with promises

to improve the pharmacology of current cancer therapeutics [3]. Drug-loaded nanoparticles (NPs) offer

several advantages over free drugs. This includes their capabilities to encapsulate large amounts of

chemotherapeutic agents, a prolonged half-life, and an increased targeting efficiency leading to reduced

toxicity [4]. Nanoparticle size, core properties, and surface modifications are considered as key factors

affecting the clearance rate and organ distribution [5]. A significant number of nanomaterials, including

inorganic nanoparticles, liposomes, micelles, and polymers have been developed for targeted drug

delivery in cancer [6, 7]. Dendrimers are considered a promising alternative because of their advanced

capabilities, which make them ideal tools for targeted drug delivery. As an example, they can increase

the solubility and bioavailability of hydrophobic drugs [8]. Further, their architecture and availability of

active functional groups allow the easy addition of moieties and the encapsulation of drugs [8].

Different types of active tumor-targeted ligands, such as RGD peptide, folic acid, and transferrin,

have been employed in conjunction with NPs to specifically direct and enhance drug accumulation in

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

224

the tumor [9]. In particular, doxorubicin-loaded HPMA-copolymer conjugated with GE11 oligopeptide

have been employed to specifically target colorectal cancer (CRC) cells with EGFR overexpression. A

high cellular uptake of the doxorubicin-loaded NP was observed, achieving a selective release of the

drug [10]. Similarly, a multifunctional nanosystem based on methotrexate-loaded guar gum NPs

functionalized with folic acid (MTXFA-GGNP) has been employed to release methotrexate at colonic pH

values (6.8) in CRC. Interestingly, the engineered NPs displayed preferential in vivo uptake in colon

tissue [11]. However, most of these approaches were mainly studied in 2D environments. In addition,

these studies also lacked a visual representation and analysis of NPs internalization in a co-culture

scenario, and importantly, their validation using native-like 3D settings.

Despite all the efforts invested in the development of targeted drug delivery systems for improved

cancer therapeutics, their performance is still limited for medical applications. New approaches are

needed to target efficiently and precisely the cancer cells. In this regard, the YIGSR laminin receptor

binding peptide has demonstrated to be a promising class of targeting ligand for specific drug delivery

in cancer treatment. Cancer cells from various sources express high levels of the 67 kDa laminin

receptor (67LR) [12], including bile duct carcinoma, colorectal carcinoma, cervical cancer, and breast

carcinoma [13], which are shown to be correlated with tumor invasiveness [14]. Earlier work has

demonstrated that the laminin binding capacity of plasma membranes isolated from human invasive

breast carcinomas is 50-fold greater than that of membranes from benign lesions or normal tissue[15].

In this study, we hypothesize that the peptide sequence YIGSR grafted onto

carboxymethylchitosan/poly(amidoamine) (CMCht/PAMAM) dendrimer NPs promotes specific and

targeted internalization into CRC cells [16]. To this end, we first characterize the morphology and

successful functionalization of CMCht/PAMAM and YIGSR-modified CMCht/PAMAM dendrimer NPs by

AFM, TEM, DLS and H1NMR. The selectivity of the YIGSR-CMCht/PAMAM dendrimer nanoparticles was

first validated in vitro using a micropatterned array of 67kDa laminin receptor. Next, we investigate the

effect of both types of dendrimer NPs on the viability of HCT-116 CRC cells or L929 fibroblasts. Next,

the targeted internalization of the NPs is assessed in a co-culture of HCT-116 CRC cells and L929

fibroblasts both in traditional well plates (2D) or encapsulated within a 3D Matrigel® matrix.

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

225

VIII-3. MATERIALS AND METHODS

VIII-3.1. Synthesis of the CMCht/PAMAM dendrimer nanoparticles

Dendrimer NPs where fabricated as described elsewhere [16]. Briefly, by using a chemical

modification route of chitin (Sigma, Germany), we synthesized CMCht with a degree of deacetylation of

80 % and degree of substitution of 47 %. CMCht/PAMAM dendrimer NPs were prepared in a step-wise

manner as follows: first, the increase of the dendrimers’ generation was carried out. An appropriate

volume of PAMAM-CT (G1.5) in methanol was transferred to a volumetric flask. The starting compound

was re-dissolved in ultrapure water to give a final concentration of 10 mg/mL and the pH was adjusted

to 6.5 with dilute hydrochloric acid solution (Riedel de-Haen, Germany). 1-ethyl-(3-dimethylaminopropyl)

carbodiimide hydrochloride (EDC, Fluka, Portugal) was then added to the solution at a molar ratio

sufficient to modify the carboxylate residue of the dendrimers. The solution was kept under agitation for

30 minutes at room temperature (RT). Ethylenediamine (EDA, Sigma, Germany) was added to the

solution at a molar ratio equal to that of EDC and left to react for 4 hours. After this period the excess of

EDC was removed by dialysis (cellulose tubing, benzoylated for separating compounds with a MW ≤

1200, Sigma, Germany). After preparing the PAMAM-amine terminated compound, an exhaustive

alkylation of the primary amines (Michael addition) was performed. An appropriate volume of PAMAM-

amine terminated was mixed with methanol (Sigma, Germany) and methyl methacrylate.

CMCht/PAMAM dendrimer NPs were then precipitated after addition of an appropriate volume of a

saturated Na2CO3 (Sigma, Germany) solution and cold acetone (Pronalab, Portugal). Precipitates were

collected by filtration and dispersed in ultrapure water for dialysis over a period of 48 hours. Finally,

CMCht/PAMAM dendrimer NPs were obtained by freezing the solution at -80 °C and freeze-dried

(Telstar-Cryodos-80) for 7 days.

VIII-3.2. Peptide attachment onto CMCht/PAMAM dendrimer nanoparticles

The attachment of YIGSR peptide (Genscript Corp, Piscataway, NJ, USA) to the NPs was performed

via carbodiimide chemistry to form an amide bond between the carboxyl groups on the NPs and the

amine groups at the YIGSR N-terminal end group. CMCht/PAMAM 1 % wt. was prepared in buffer 2-(N-

morpholino)ethanesulfonic acid (MES), Sigma Aldrich, Germany), and mixed with the appropriate

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

226

amount of NHS-Sulfo, EDC and a 10 mg/mL YIGSR peptide solution (dissolved in MES buffer). The

solution was stirred for 24 hours and the NPs were recovered by dialysis against ultrapure water for 48

hours followed by freeze-drying process for 7 days.

VIII-3.3. Labelling of CMCht/PAMAM dendrimer nanoparticles with FITC

The conjugates of CMCht/PAMAM-FITC were obtained by covalently bonding the CMCht amine

group and the isothyocianate group of FITC (10 mg/mL) (Sigma, Germany) in anhydrous dimethyl

sulfoxide (DMSO, Norconcessus, Portugal) creating a thiourea bond. First, a 10 mg/mL CMCht/PAMAM

dendrimer NPs solution was prepared in a carbonate-bicarbonate coupled buffer at pH 9.2. Then, a

solution of FITC/DMSO was added under agitation and kept in the dark at 4 °C for 8 hours. Next, the

FITC-labeled CMCht/PAMAM dendrimer NPs solution was dialyzed against ultrapure water for 24 hours

to remove unlinked FITC and filtered (pore size < 220 nm) in sterile and dark conditions. The final

product was obtained after freeze-drying.

VIII-3.4. Gemcitabine incorporation into CMCht/PAMAM dendrimer nanoparticles

Gemcitabine (GEM) was incorporated in the NPs by mixing an aqueous solution of YIGSR-

CMCht/PAMAM dendrimer NPs with an aqueous GEM solution with a final concentration of 5x104 M.

Saturated sodium carbonate solution (Na2CO3) and acetone were then added to the mixture. The

resulting precipitates were dialyzed against ultrapure water. GEM-loaded YIGSR-CMCht/PAMAM

dendrimer NPs were obtained by freezing the final solution at -80 ºC and freeze-drying the samples.

VIII-3.5. Characterization of the CMCht/PAMAM dendrimer nanoparticles

VIII-3.5.1. Protein quantification

The concentration of YIGSR peptide grafted onto the dendrimer NPs was estimated by UV

spectroscopy (between 200 nm and 275 nm) and the Micro-BCA (562 nm) protein quantification assay

using a microplate reader (BIO-TEK, EUA). Samples of CMCht/PAMAM and modified YIGSR-

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

227

CMCht/PAMAM dendrimer NPs were dissolved in ultrapure water at a final concentration of 1 % and

compared to standards for protein quantification in triplicates.

VIII-3.5.2. Dynamic light scattering (DLS)

Zeta potential and the size of the CMCht/PAMAM and YIGSR-CMCht/PAMAM dendrimer NPs were

measured in a particle size analyzer (Zetasizer Nano ZS, Malvern Instruments, UK). Particle size

analyses were performed by dynamic light scattering (DLS) in phosphate buffered saline (PBS) at pH

7.4 with low concentration of NPs and using disposable sizing cuvettes. Electrophoretic determinations

of Zeta potential were investigated using the universal ‘dip’ cell at pH 7.4 in PBS solution. Zeta potential

was also investigated in water.

VIII-3.5.3. Nuclear magnetic resonance spectroscopy (NMR)

The structure of CMCht/PAMAM and YIGSR-CMCht/PAMAM dendrimer NPs was determined by

H1NMR. Both types of NPs were dissolved in deuterated water (D2O). Then, the NMR spectra were

obtained with a Mercury -400BB operating at a frequency of 399.9 MHz at 50 °C. The one-dimensional

1H spectra were acquired using a 45° pulse, a spectral width of 6.3 kHz and an acquisition time of

2.001

VIII-3.5.4. Scanning Transmission Electron Microscopy (STEM)

Dendrimer NPs morphology was investigated by STEM (Philips CM-12, FEI, The Netherlands),

equipped with a MEGA VIEW-II DOCU camera and Image Software Analyzer SIS NT DOCU. For that, the

NPs were stained with 2 % of phosphotungstic acid and placed on the copper grids for observation,

VIII-3.5.5. Fourier Transform Infrared (FTIR)

The dried dendrimer NPs, prior to, and after YIGSR attachment, were mixed with potassium

bromide (KBr) and placed between two disks. Pressure was applied (4.5 metric tons, Carver Inc) on the

disk to form a pellet. The spectrum of the formed pellet was measured by FTIR spectroscopy (Exuinox

55, Bruker Optics Inc., USA) in the range of 400 and 3500 cm-1.

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

228

VIII-3.5.6. Microcontact printing

Poly(dimethylsiloxane) (PDMS) 1:10 (v/v), Sylgard 184; Dow Corning) stamps were replicated from

an SU-8 mold (MicroChem) fabricated by UV-photolithography (MIDAS MDA-400MA). The PDMS stamp

was rendered hydrophilic by plasma treatment to maximize 67 LR patterning. Glass coverslips were

plasma treated for 1 minute and incubated for 1 hour under anoxic conditions in anhydrous methanol

solution of 5 % (v/v) 3-aminopropyl) triethoxysilane (APTES). Next, the samples were washed with

methanol followed by ultrapure water. Then, APTES-functionalized coverslips were incubated for 30

minutes in a solution of 1 % (v/v) gluteraldehyde in ultrapure water. The micropatterned PDMS stamp

was incubated with 10 μg/mL LR protein (Abcam, Portugal) solution for 2 hours and excess was

removed by nitrogen drying. Immediately, the PDMS stamp containing the laminin receptor was placed

in contact with the functionalized glass coverslip for 5 minutes. After releasing the stamp, FITC-labeled

YIGSR-CMCht/PAMAM dendrimer NPs were incubated overnight.

VIII-3.6. In vitro cell culture in 2D and 3D platforms

VIII-3.6.1. Cell culture

Human HCT-116 colorectal cancer cells and L929 fibroblast were obtained from the American

Collection of Cell Cultures (ATCC, USA). Cells were grown in Dulbecco’s Modified Eagle Medium (Sigma

Aldrich, Germany) supplemented with 10 % (v/v) fetal bovine serum and 1 % (v/v)

penicillin/streptomycin (Gibco, USA) at 37 °C and 5 % CO2. Subcultures of cells were performed when

confluence reached values of ~90 %. Then, cells were plated at an initial density of 1x104 cells per well

in a 24-well plate for metabolic activity and DNA quantification. Co-cultures were used for internalization

studies (1x104 of each cell type in 24 well plates).

VIII-3.6.2. Immunocytochemistry

L929 were pre-stained in red with CellTracker™CM Dil (1 µM) and seeded at an initial density of

1x104 cells per well in a 24-well plate. At days 1 and 3, HCT-116 cancer cells and L929 fibroblasts in

co-culture were fixed with 4 % formalin (Sigma Aldrich, Germany) for 10 minutes and washed with PBS.

Cells were permeabilized with 0.1 % (v/v) Triton X-100 (Sigma Aldrich, Germany) in PBS for 5 minutes

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

229

and blocked with 3 % BSA (Sigma Aldrich, Germany) in PBS for 30 minutes. Immunolabeling was

performed using a rabbit anti-human monoclonal antibody against 67 kDa LR (Abcam, 1:200) as

primary antibody, prepared in PBS 1 % BSA and incubated overnight at 4 ºC. The samples were

incubated for 1 hour at RT with the respective secondary fluorochrome-conjugated antibodies, anti-

rabbit/mouse IgG (Invitrogen, dilution 1:1000). Immunofluorescence was observed under transmitted

fluorescence microscopy (Zeiss, Germany) (LR in green: ex/em 488/517; Cell Tracker™CM Dil in red:

ex/em 594/618).

VIII-3.6.3. Metabolic activity

Alamar blue (AB) assay was performed according to supplier’s instructions to assess the metabolic

activity of cells. The effect of CMCht/PAMAM, YIGSR-CMCht/PAMAM and GEM-YIGSR-CMCht/PAMAM

dendrimer NPs was assessed in terms of metabolic activity in each cell type separately. After 24 hours

and 72 hours in culture, specific cell culture medium containing 20 % (v/v) AB was added to the

different culture wells. The system was incubated for 4 hours after which fluorescence was monitored at

590 nm emission wavelength (excitation wavelength 530 nm), using a microplate reader (FL 600, Bio-

Tek Instruments). After each AB determination, cells were rinsed with PBS and fresh culture medium

was added to continue the culture. Medium with 20 % (v/v) AB was used as a blank.

VIII-3.6.4. DNA quantification

To assess the effect of CMCht/PAMAM, YIGSR-CMCht/PAMAM and GEM-YIGSR-CMCht/PAMAM

dendrimer NPs on the cancer cells and fibroblasts separately, the total amount of DNA was measured

using the fluorimetric double-stranded DNA (dsDNA) quantification kit (Quant-iT™ PicoGreen® dsDNA

Assay Kit, Molecular Probes). The cells were washed with sterile PBS and lysed with 1 mL of ultrapure

water in each well. The cellular suspensions were transferred into microtubes, incubated in a water

bath at 37 °C for 1 hour, and stored at -80 °C for further analysis. DNA standards were prepared with

concentrations varying between 0 and 2 g/mL. Next, 28.7 L of sample (or DNA standard), 71.3 L

of PicoGreen and 100 L of TE buffer were mixed in each well of an opaque 96-well plate (Alfagene,

Portugal), according to the manufacturer’s protocol. Next, after 10 minutes incubation at RT,

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

230

fluorescence was measured using an excitation and emission wavelengths of 485 nm and 528 nm,

respectively. Sample DNA values were read off from DNA standard curve.

VIII-3.6.5. Live/dead assay

The cell viability was assessed by calcein-AM and propidium iodide (PI; Life Technologies,

Carlsbad, CA, USA) staining. L929 cells were first stained with CellTracker Blue CMAC (Alfagene,

Portugal), whereas HCT-116 cells were left unstained. After 72 hours, three samples were incubated in

1 µg/mL calcein-AM and 5 µg/mL PI prepared in DMEM low glucose supplemented with sodium

bicarbonate and 1 % (v/v) antibiotic/antimycotic solution for 30 minutes in the dark at 37 ºC in the 5 %

CO2. After washing in PBS, samples were immediately analyzed in a confocal microscopy (Calcein-AM in

green: ex/em 495/515 nm; PI in red: ex/em 495/635 nm; CellTracker Blue CMAC in blue: ex/em

353/466 nm) (Leica, SP8, Germany).

VIII-3.6.6. Cellular uptake

HCT-116 colorectal cancer cells and L929 fibroblasts were pre-stained in Far Red and in Blue

CMAC CellTracker (Alfagene, Portugal), respectively. After 30 minutes incubation at 37 ºC, cells were

seeded in co-culture at a density of 1x104 cells (each type)/well in 24 well plates, and allowed to adhere

overnight. Then, the culture medium was replaced by fresh one supplemented with FITC-

CMCht/PAMAM and FITC-YIGSR-CMCht/PAMAM dendrimer NPs at a final concentration of 0.5 mg/mL.

At 24 hours and 72 hours, cells were washed with PBS, fixed with 4 % (v/v) formalin for 20 minutes at

RT, washed again 3 times with PBS, and permeabilized with 0.2 % (v/v) Triton X-100 in PBS for 10

minutes. Next, cells were washed with PBS. The cell nuclei were stained with DAPI and F-actin with

Phalloidin, as described above. The cells were imaged by confocal microscopy to assess the

internalization of NPs (FITC – 488 nm). For 3D studies, the same procedure was used. In this case, the

cells were encapsulated into a 50 µL of Matrigel matrix (Enzifarma, Portugal).

VIII-3.6.7. Flow cytometry analysis

For internalization analysis, the cells were co-cultured in medium supplemented with 0.5 mg/mL

of FITC-labeled YIGSR-CMCht/PAMAM dendrimer NPs in 24-well plates (1x104 cells; each type/well).

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

231

Cells cultured in complete culture medium were used as controls. After 24 hours and 72 hours, the

cells were trypsinized (TrypleX, Life technologies, USA) and a cell strainer was used to avoid cell

clusters. Next, 0.5 mL of DMEM complete medium was added to each well and the samples were

transferred to flow cytometry tubes. To distinguish cells in co-culture, 10 µL of PE-CD44 antibody

(Abcam, Portugal) was incubated with HCT-116 cells for 20 minutes at RT. After washing with 2 mL of

2 % (v/v) FBS in PBS, the tubes were centrifuged at 300 rpm for 5 minutes and re-suspended in 1 %

(v/v) formalin in PBS. The cell suspensions were analyzed in FACSCalibur flow cytometer (BD

Biosciences, USA). A similar approach was followed for cell viability studies. In this case, the tubes were

centrifuged at 300 rpm for 5 minutes and HCT-116 were stained with FITC-CD44 (Abcam, Portugal) for

30 minutes and re-suspended in 2 % (v/v) FBS in PBS. Then, 1 µL of PI were added to suspensions

and incubated for 10 minutes at RT, after which they were analyzed by FACSCalibur flow cytometer. For

internalization and cell viability assays in 3D, cells encapsulated in Matrigel were retrieved using 500 µL

of Corning Cell Recovery Solution (Corning, USA) for 30 minutes at 4 ºC and the subsequent protocol

was performed as previously described.

For all assays, calibrate beads three-color kit (BD CaliBRITETM beads, USA) was used to adjust the

instrument settings before samples are run on the flow cytometer. Finally, data were analyzed using the

Flowing Software 2.

VIII-3.6.8. Statistical analysis

Statistical analysis was performed using GraphPad Prism 5.0 software version 5.0a. The non-

parametric Mann–Whitney test was used to compare two groups, whereas comparison between more

than two groups was performed using the Kruskal–Wallis test followed by Dunn’s comparison test. A

value of p < 0.05 was considered statistically significant.

VIII-4. RESULTS AND DISCUSSION

VIII-4.1. Physico-chemical characterization of dendrimer nanoparticles

We first investigated the attachment of the YIGSR peptide on the dendrimer NPs surface via

carbodiimide chemistry (Figure VIII-1A), and the morphology and size of both type of NPs. STEM images

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

232

revealed a monodisperse round-shaped morphology (Figure VIII-1B), which was confirmed by AFM

measurements (see Supplementary Information VIII-S1A). These results suggested the successful

grafting of the peptide sequence, which conferred the YIGSR-CMCht/PAMAM dendrimer NPs a larger

size. DLS was used to quantify the size of the NPs. Figure VIII-1C shows the peaks associated with both

the CMCht/PAMAM and YIGSR-CMCht/PAMAM dendrimer NPs. In the former, two main peaks were

obtained, which corresponded to individual and aggregated NPs. The first peak provided the average

size of the individual NPs was 54 ± 7 nm, whereas the second one corresponded to NP aggregates with

a much bigger size. In contrast, YIGSR-CMCht/PAMAM dendrimer NPs displayed three peaks. The first

peak located around 6 nm corresponded to unreacted PAMAM G1.5. The reason is that the first step in

the synthesis of the dendrimer NPs is increasing the generation of the PAMAM-CT (G 1.5), which has a

theoretical size equal to the one provided by the peak, and therefore, this peak was assumed to

correspond to unreacted G1.5. The second peak located around 130 ± 20 nm corresponds to the

YIGSR-modified dendrimer NPs. Finally, the last peak is associated with aggregated NPs. Altogether,

these results show that attachment of the peptide significantly increased the size of the NPs by ~2.5-

fold without affecting the NPs polydispersity index (PDI) with values of 0.29 and 0.37 for

CMCht/PAMAM and YIGSR-CMCht/PAMAM PDI, respectively.

Figure VIII-1D shows the FTIR spectrum of the CMCht/PAMAM and the newly modified YIGSR-

CMCht/PAMAM dendrimer NPs. The spectrum of the modified NPs shows an increase in peak intensity

at 1700 cm-1 attributed to the amide II and C-N stretching vibrations [17]. This is an evidence of the

presence of amine groups as well as of the chemical bonding between the CMCht/PAMAM and the

peptide. Similarly, the 520 cm-1 free amino acid peak and the 1120 cm-1 NH3 vibration appear after NP

modification [12, 18]. Moreover, both NPs showed characteristic peaks, such as the strong and broad

peak of absorption at ~3400 cm-1, which are assigned to the stretching vibration of the structural N-H

overlapping the O-H stretching. Similarly, the attachment of the YIGSR peptide was further confirmed by

the analysis of the chemical signature obtained by H1NMR (Figure VIII-1E) [19]. Bearing in mind the

analysis of H1NMR spectra (see Supplementary Information VIII-S2), we can state that the signature

protons of CH2 and CH3 of the YIGSR peptide are present in the YIGSR-CMCht/PAMAM around 2 and 3

ppm, indicating the successful grafting of the peptide. Moreover, the broad singlet at 3.02 ppm of the

CMCht/PAMAM is also present in the modified nanoparticles. Taken together, the results obtained

indicate that the peptide was successfully conjugated to the NPs.

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

233

We next measured the Zeta potential values of both unmodified CMCht/PAMAM and YIGSR-

modified CMCht/PAMAM dendrimer NPs. Figure 1 F shows that the formers have a value of -14.7 ±

3.0 mV in PBS and -34.3 ± 3.0 mV in ddH20. In contrast, YIGSR-modified CMCht/PAMAM dendrimer

NPs have a value of -9.3 ± 0.8 mV in PBS and -19.3 ± 0.8 mV in ddH20. The difference between YIGSR-

modified dendrimer NPs (less negative) and unmodified NPs (more negative), indicates a reduction in

the number of carboxylic acid groups on the NPs surface following peptide binding. Moreover, the

addition of the positively-charged Arginine present in the peptide sequence suggests a successful

grafting. It is expected that CMCht has a good pH and ion sensitivity in aqueous solution due to

abundant –COOH and –NH2 groups. On the other hand, since the Zeta potential of NPs is determined

by their surface properties, this data shows the negatively charged carboxymethyl groups in the

CMCht/PAMAM dendrimer NPs. The different Zeta potential values obtained in ddH20 and PBS can be

explained by the fact that the CMCht behaves as a weak polyanionic polyelectrolyte at physiological pH,

i.e., the amino groups are not protonated and most of carboxyl groups are not dissociated.

Further characterization was achieved by quantifying the amount of coupled peptide using the BCA

protein quantification assay and compared to a standard curve using BSA (Figure VIII-1G). Results show

a significant difference between the levels of protein on non-modified and YIGSR-grafted CMCht/PAMAM

dendrimer NPs. The former shows an amount of 3.60 ± 1.62 µg/mL, whereas the later has a

concentration of 71.00 ± 3.75 µg/mL. This indicates that the peptide sequence YIGSR is successfully

attached to the dendrimer NPs. Finally, the presence of the peptide on the CMCht/PAMAM dendrimer

NPs was investigated by UV spectroscopy (Supplementary Information VIII-S1B). The peptide sequence

YIGSR incorporates the amino acid tyrosine, which exhibits two characteristic absorbance peaks in the

UV spectral region at approximately 230 nm and 280 nm [20]. The UV spectra were obtained using

CMCht/PAMAM and modified YIGSR-CMCht/PAMAM dendrimer NPs for comparison. Two peaks

located around 275 nm were observed. This indicated that the peptide is effectively grafted to the

polymer backbone.

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

234

Figure VIII-1 - Physico-chemical characterization of dendrimer nanoparticles. (A) Schematic representation of the dendrimer

NPs modification with YIGSR peptide from laminin via carbodiimide chemistry. (B) STEM image of

CMCht/PAMAM and YIGSR-CMCht/PAMAM dendrimer NPs. (C) Representative graphic of the particle size

distribution of CMCht/PAMAM and YIGSR-CMCht/PAMAM NPs, showing that the peptide sequence

conferred the NPs a larger size. (D) FTIR analysis of CMCht/PAMAM and YIGSR-CMCht/PAMAM dendrimer

NPs, showing an increase in intensity of the bands at 1700 cm-1 for YIGSR-CMCht/PAMAM dendrimers. (E)

H1NMR spectra of YIGSR-CMCht/PAMAM, CMCht/PAMAM, and YIGSR. The YIGSR-CMCht/PAMAM shows a

peak at 3 ppm and no peak at 4.3 ppm suggesting that the substitutions occurred mainly on the C2 amino

group of CMCht. (F) Table of the mean NP diameter and Zeta potential in PBS and ddH 20, showing an

increase in diameter and a decrease in Zeta potential upon the addition of YIGSR. (G) Determination of

peptide concentration by protein quantification (BCA protein quantification) in CMCht/PAMAM and YIGSR-

CMCht/PAMAM.

VIII-4.2. Targeted detection of FITC-YIGSR-CMCht/PAMAM on a micropatterned laminin receptor array

The 67 kDa high affinity LR is a non-integrin cell surface receptor of the extracellular matrix. The

expression of this receptor is thought to be increased in neoplastic cells, which directly correlates with

an enhanced invasiveness and metastatic potential [21]. To confirm this higher expression, we first

performed an immunocytochemistry analysis to assess qualitatively the differences in LR expression in

HCT-116 and L929 cells at 24 h and 72 h. HCT-116 cells displayed round morphologies and

proliferated forming dense aggregates (Figure VIII-2A). In contrary, L929 fibroblasts showed highly

spread phenotypes. Importantly, HCT-116 displayed a large LR signal both at 24 hours and 72 hours.

This is in contrast to L929 fibroblasts, which showed a low LR expression. Taken together, these results

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

235

suggest that the 67 kDa LR may be employed as a cancer-specific target to specifically deliver drug-

loaded YIGSR-CMCht/PAMAM dendrimer NPs in a precise and efficient manner.

Next, we evaluated in vitro the specificity of the YIGSR-CMCht/PAMAM dendrimer NPs toward the

LR. For this, an array of 67 kDa LR was micropatterned on a functionalized glass coverslip by micro-

contact printing to recreate in vitro the physiological target (Figure VIII-2B). Next, FITC-labeled YIGSR-

CMCht/PAMAM dendrimer NPs were incubated on top of the LR array. We observed that the FITC-

labelled NPs attached mainly on top of the triangular micropatterned regions, which contained the 67

kDa LR protein. Overall, these results show a strong affinity of between the NPs grafted with a portion of

YIGSR peptide and the LR protein. However, more complex experiments involving protein receptors

other than LR may need to be also tested to univocally prove a really specific affinity for this receptor.

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

236

Figure VIII-2 - Specificity of FITC-YIGSR-CMCht/PAMAM dendrimer NPs on a micropatterned LR array. (A)

Immunocytochemistry of anti-67kDa LR (green) performed on a co-culture of HCT-116 CRC cells and L929

fibroblasts (stained in red; See supplementary Information S3) cells at 24 h and 72 h. An overexpression of

LR on HCT-116 is visible at both time points. (B) (Left) Schematic representation of the steps involved in the

microcontact printing of the LR: (i) incubation of the PDMS stamp with the LR solution; (ii) microcontact

printing of the LR on top of a functionalized (APTES 5 % and glutaraldehyde) glass coverslip; (iii) PDMS

stamp removal and rinsing with ddH2O; (iv) incubation with FITC-YIGSR-CMCht/PAMAM; and (v) rinse with

ddH2O to remove the excess of FITC-YIGSR-CMCht/PAMAM. (Right) Representative brightfield and

fluorescence microscopy images of the PDMS stamp and FITC-YIGSR-CMCht/PAMAM NPs array displaying

the triangular micro-sized LR features. Scale bar: 100 μm.

VIII-4.3. Cytotoxicity tests: metabolic activity and cell proliferation

The cytotoxicity of the dendrimer NPs was assessed by measuring cell metabolic activities and

proliferations by means of Alamar blue and DNA quantification assays. HCT-116 cancer cells and L929

fibroblasts were exposed to CMCht/PAMAM, YIGSR-CMCht/PAMAM, and GEM-YIGSR-CMCht/PAMAM

at a final concentration of 0.5 mg/mL over a period of 72 hours. This value was selected taking into

consideration the maximum concentration of pure dendrimer NPs (i.e. non-modified) that may be

employed without observing any cytotoxic effect [16]. It is important to notice that both the metabolic

activity and cell viability were assessed in monoculture to discard the possibility of inherent toxicities of

non-modified and modified dendrimer NPs. Both HCT-116 cancer cells and L929 fibroblasts

significantly proliferated from day 1 to day 3 in the absence (control) and presence of CMCht/PAMAM

and YIGSR-CMCht/PAMAM NPs (Figure VIII-3). The obtained results showed no deleterious effects of

the non-modified CMCht/PAMAM and the YIGSR-CMCht/PAMAM dendrimer NPs on cell viability and

proliferation in both cell types. This indicates that the grafting of the YIGSR sequence did not

significantly affect cellular behavior.

Conversely, GEM-loaded YIGSR-CMCht/PAMAM NPs had a clear cytotoxicity in both cell types due

to the release of the drug, as demonstrated by the decrease in the metabolic activity and DNA amount

(Figure VIII-3A and Figure VIII-3B). Interestingly, this effect was mostly observed at early time points in

HCT-116 cancer cells in contrast to L929 fibroblasts, where the cytotoxicity was only observed at 72

hours. This suggests that the drug was internalized faster in HCT-116 cancer cells due to LR

overexpression. For L929 fibroblasts, viability was most likely affected by the release of GEM from the

NPs. Regarding the release of GEM from the NPs, and its potential in vivo side effects, it is plausible

that the NPs release some amount of drug in the medium before it is internalized by the cells. However,

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

237

its effect should not be significant because the same conditions and time-points were tested in co-

culture (similar to in vivo scenario) with an increased cell death for HCT-116 cancer cells. Overall, we

do not anticipate any significant adverse effect of the released drug, within the range of the used

concentrations, on non-cancerous cells when eventually used in vivo. Moreover, it is worth noticing that

these effects were observed when HCT-116 and L929 were cultured in separate, and not in co-culture.

Only after assessing the non-cytotoxic effects of the peptide modification in mono-culture, its effect in

co-culture was investigated. Overall, these results indicate that GEM-loaded YIGSR-CMCht/PAMAM NPs

affected the viability and proliferation of HCT-116 cancer cells and L929 fibroblasts in contrast to non-

modified and peptide-grafted dendrimer NPs, but in different time scales.

Figure VIII-3 - Cytotoxicity studies. (A) Metabolic activity determined by Alamar blue, and (B) DNA quantification results of

HCT-116 and L929 cells in the presence of 0.5 mg/mL of CMCht/PAMAM, YIGSR-CMCht/PAMAM, and

GEM-loaded YIGSR-CMCht/PAMAM, at 24 hours, 48 hours and 72 hours. *Significant differences when

comparing the different conditions with control (0 mg/mL) at each time point. # significant differences were

observed when comparing to CMCht/PAMAM at each time point. $ Significant differences were observed

when comparing to YIGSR-CMCht/PAMAM at each time point. P < 0.05.

VIII-4.4. Targeted internalization of dendrimer nanoparticles in 2D and 3D

We next investigated whether the attachment of YIGSR on the surface of the dendrimer NPs

influenced their preferential uptake by HCT-116 cancer cells, which overexpress 67 kDa LR at their

surface, in a co-culture with L929 fibroblasts (see Figure VIII-4A). Nanoparticle internalization was first

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

238

investigated in 2D. As expected, FITC-CMCht/PAMAM NPs widely internalized in both cell types at 24

hours and 72 hours, and localized in the cytoplasm and around the nucleus (Figure VIII-4B) [16, 19]. In

contrast, the peptide-modified YIGSR-CMCht/PAMAM dendrimer NPs internalized preferentially into

HCT-116 cancer cells (red) at 24 hours when seeded in a co-culture with L929 fibroblasts (blue).

Importantly, a qualitative increase in the number of internalized NPs could be observed from 24 hours

to 72 hours. This suggested that peptide-grafted NPs were specifically targeted towards HCT-116 cells

overexpressing 67 kDa LR.

We next used an in vivo-like 3D scenario to determine the efficiency of the cellular uptake mediated

by the targeted interaction of the YIGSR peptide with the LR. For this, we encapsulated HCT-116 cancer

cells and L929 fibroblasts into a 3D matrix of Matrigel®. Figure VIII-4B (bottom) shows the cells cultured

in the absence (control) and presence of CMCht/PAMAM and YIGSR-CMCht/PAMAM dendrimer NPs.

We first investigated whether the NPs were capable to diffuse across the meshes of the hydrogel and

target the HCT-116 cells. After 24 hours no significant internalization of both CMCht/PAMAM and

YIGSR-CMCht/PAMAM NPs was observed (data not shown). Instead, the NPs aggregated in the outer

membrane of the cells (see Supplementary Information VIII-S1E). This phenomenon may be a

consequence of NPs diffusing through the dense mesh of the crosslinked hydrogel, and therefore,

taking longer than in 2D to interact with cells and internalize. In contrast, after 72 hours, FITC-labeled

NPs spread all across the 3D hydrogel accumulating both inside and around the HCT-116 cells.

Notably, it was found that YIGSR-CMCht/PAMAM NPs internalized preferentially into the HCT-116

cancer cells (Figure VIII-4B1), white arrowheads). In contrast, L929 fibroblasts did not show any

significant evidence of NPs internalization (see Figure VIII-4B2)). These results highlight the differences

between 2D and 3D scenarios in drug delivery applications, even in terms of simple biological

processes such as the internalization of NPs.

We next quantified the amount of NPs that were internalized in both conditions (2D vs 3D). To this

end, flow cytometry analysis was performed using the same experimental conditions. HCT-116 cancer

cells were stained with an anti-CD44 antibody-PE (see Supplementary Information VIII-S1C). In this

case, only the HCT-116 cells were stained, serving as a differential marker. The obtained results were in

agreement with the fluorescence microscopy results. In 2D, a maximum internalization of NPs (96.27

%) for the HCT-116 cells was observed at 24 hours, compared to the low internalization (2.97 %) in

L929 fibroblasts (Figure VIII-4C, left). At 72 hours, the internalization of NPs increased slightly (98.45 %

for HCT-116 cancer cells and 6.56 % for L929 fibroblasts). This clearly showed the lower levels of

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

239

internalization in L929 fibroblasts when compared to those of HCT-116 cancer cells. In contrast, in 3D

scenario, we found that the overall levels of internalization were not as high as in 2D (Figure VIII-4C,

right). After 24 hours, the level of internalization of targeted NPs in HCT-116 cancer cells and L929

fibroblasts was 65.52 % and 3.54 %, respectively. At 72 hours, these values increased up to 85.44 %

and 9.95 %, respectively. As mentioned previously, these differences between 2D and 3D may be

related to the fact that in 3D dendrimer NPs need to diffuse through the dense hydrogel mesh to target

the encapsulated cells.

Together, these results indicate a targeted internalization of YIGSR-CMCht/PAMAM dendrimer NPs

towards HCT-116 cells overexpressing 67 kDa LR both in 2D and 3D. However, different levels of

internalization levels were obtained when comparing both conditions, reinforcing that traditional 2D

models may poorly correlate with human disease states.

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

240

Figure VIII-4 - Internalization of dendrimer nanoparticles. A) Schematics of the different conditions studied. B) Confocal

microscopy images of HCT-116 cancer cells (red) and L929 fibroblasts (blue) co-culture in the presence of

YIGSR-CMCht/PAMAM and CMCht/PAMAM dendrimer NPs in (top) 2D for 24 hours and 72 hours, and

(bottom) in 3D for 72 hours. In the latter, (1) and (2) show a higher magnification of HCT-116 and L929

cells, respectively, showing the internalization of YIGSR-CMCht/PAMAM dendrimer NPs exclusively for HCT-

116 (observed only with the 488 nm filter). Inset images show a more detailed view of the NPs (indicated by

white arrowheads). (C) Flow cytometry results of NPs internalization in co-culture in both 2D and 3D for 24

hours and 72 hours. Scale bars: 20 µm.

VIII-4.5. Targeted gemcitabine delivery using dendrimer nanoparticles in 2D and 3D

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

241

After verifying the targeted and preferential internalization of YIGSR-CMCht/PAMAM NPs towards

HCT-116 cancer cells, we next investigated the gemcitabine (GEM) release from YIGSR-CMCht/PAMAM

in a co-culture of HCT-116 cancer cells and L929 fibroblasts in 2D and 3D [22]. In a previous work, we

demonstrated an encapsulation efficiency of GEM into CMCht/PAMAM NPs of 52.99 % ± 3.50 %, with a

drug-loading efficiency of 9.42 % ± 3.50 % [23]. First, the viability of cells was analyzed by Live/Dead

assay in a 2D co-culture at 24 hours and 72 hours in the presence of GEM-YIGSR-CMCht/PAMAM

(Figure VIII-5A). The results show that L929 fibroblasts (in blue) were not substantially affected by the

GEM-YIGSR-CMCht/PAMAM NPs; they were co-stained in blue and in green (live cells). In contrast, HCT-

116 cancer cells (unstained) displayed a larger mortality at 24 hours (red - dead cells) (Figure VIII-5A)

indicating a specific GEM release inside these cells. It is worth noticing that most of the cancer cells

were alive at 24 hours. However, they displayed an elongated – perturbed – phenotype with the

appearance of blebs, which suggested the beginning of the apoptotic process [24]. In contrast, this was

not observed in L929 cells. This was more stringent at 72 hours, where mainly HCT-116 cancer cells

were dead. In contrast, L929 were mostly alive, displaying a morphology consistent with its normal

phenotype and not significantly perturbed. We confirmed these qualitative observations by quantifying

the amount of live/dead cells by flow cytometry. For this, CD44-FITC was used to label HCT-116 cancer

cells, while L929 fibroblasts were CD44 negative and PI labelled the dead cells. Figure 5 B shows a

4.70 % of cell death (PI staining) in the CD44 positive population (HCT-116) in 2D at 24 hours, and

3.44 % for the CD44 negative population (L929). Although not very substantial at such early time point,

there is a difference between both cell death values, being higher for cancer HCT-116 cells. The same

trend was observed in 3D, where cell death for the CD44 positive (HCT-116) and negative (L929)

populations was 30.16 % and 26.07 %, respectively. Note that the percentage of overall cell death is

higher in 3D than in 2D. This phenomenon could be related to the process of retrieving cells from the

hydrogel. This process consists of leaving the hydrogels in the recovery solution at 4 ºC for at least 30

minutes. This may damage the cells, and therefore, may explain the increase in percentage of cell

death.

At 72 hours, we observed a high level of internalization in HCT-116 cancer cells in 2D compared to

L929 fibroblasts. This resulted in a significantly higher level of cell death (68.72 %) observed in the

CD44 positive population (HCT-116) compared to L929 fibroblasts (3.20 %); this was in agreement with

the previously studied internalization phenomena. In 3D, after exposing the co-culture to GEM-YIGSR-

CMCht/PAMAM for 72 hours, cell death percentage of the CD44 positive population (HCT-116) was

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

242

31.15 %, while for the CD44 negative population (L929) was 19.52 %. These results show that the

intracellular release of GEM was successfully achieved mainly toward cancer cells. However, when

comparing the same condition at 24 hours, we observed that the level of cell death did not increase

significantly. This could be explained by several possible factors: (i) the GEM is released within the first

24 hours, being almost 40 % released in the first 6 hours of culture [23]. Since a static culture is

employed and fresh medium is not added, the drug levels are stable, therefore, cell death is also stable

both at 24 hours and 72 hours; (ii) internalization rates do not drastically increase from 24 hours to 72

hours in 3D. This also corroborates the fact that there is no significant difference in the amount of drug

available inside the cells to induce apoptosis from 24 hours to 72 hours; and (iii). At 72 hours cells

seem to be in the beginning of the process of apoptosis (Figure VIII-5A), as suggested by the presence

of both green and red signals. This also suggests that HCT-116 cancer cells may take longer time to

complete the cell death process and to be detected by flow cytometry.

Taken together, these results suggest the successful targeting and release of the GEM-YIGSR-

CMCht/PAMAM dendrimers NPs into the HCT-116 cells taking advantage of the 67 kDa LR

overexpression on the cell surface.

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

243

Figure VIII-5 - The cell viabilities of HCT-116 and L929 cells in the presence of GEM-YIGSR-CMCht/PAMAM dendrimers. (A)

Live/Dead assay performed on a co-culture of HCT-116 (unstained) and L929 (in blue) cells at 24 hours and

72 hours in 2D. (B) Flow cytometry analysis of cell viabilities in the presence of GEM-YIGSR-CMCht/PAMAM

in a co-culture of HCT-116 (CD44 positive) and L929 cells (CD44 negative cells).

Finally, the rationale of using YIGSR (a peptide derived from laminin) as a tumor targeting is based

on the solid evidences that the levels of expression of the 67 kDa LR directly correlate with the

invasiveness and metastatic potential in numerous tumors.[14, 15, 25]. It is our goal to further test

these newly developed NPs using primary colorectal cancer cells and non-cancerous colorectal cells in

a dynamic microfluidic system, in order to better mimic the native scenario.

Overall, the future perspective of targeted NPs for cancer research is promising. Several

formulations currently under pre-clinical and clinical studies for CRC may soon reach the market. A

future challenge common to recent technologies, with no exception for targeted NPs, is the need for

implementation of systems or protocols to determine the molecular expression profile of tumors from

patients with CRC. Its classification according to the genetic profile, stage of tumor development, and

putative targeting molecules are imperative [26]. By achieving this, the rational administration of

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

244

precise-targeted nanomedicines containing the most effective drug combination may revolutionize the

way cancer is treated.

VIII-5. CONCLUSIONS

The conjugation of ligands as peptides, aptamers and other small molecules, such as antibodies,

on the surface of nanoparticles with the purpose of cell targeting has resulted in a new generation of

nanoparticles for cancer therapy with enhanced in vitro and in vivo specificity. The present work first

demonstrates by a simple micro-pattern assay that YIGSR laminin peptide on the surface of

CMCht/PAMAM NPs links selectively to 67 kDa receptor protein. It is also demonstrated in this work

that the attachment of YIGSR peptide on the surface of CMCht/PAMAM dendrimer NPs leads to their

targeting and preferential uptake by HCT-116 cells, and consequently, releasing of the anti-cancer drug

intracellularly, when in co-culture with fibroblasts. The novel nanoparticles demonstrate that they may

be considered as a promising tool for targeted drug delivery in colorectal cancer cells and other types of

cancer, whose cells expresses high levels of the 67LR. In the future, such technology will need to be

evaluated in other CRC cell lines and primary cells to fully disclose the potential of YIGSR-targeted NPs

in cancer therapeutics. Finally, the potential of YIGSR-CMCht/PAMAM NPs for further applications is

immense. This tool may bring closer to a clinical and pharmaceutical translatable setting for continuing

research.

VIII-6. ACKNOWLEDGMENTS

This work was financially supported through the project FROnTHERA (NORTE-01-0145-FEDER-

000023), Norte Portugal Regional Operational Programme (NORTE 2020), under the Portugal 2020

Partnership Agreement, through the European Regional Development Fund (ERDF); M. C. for her PhD

scholarship NORTE-08-5369-FSE- 000044, funded by Programa Operacional Regional do Norte, Fundo

Social Europeu, Norte 2020 TERM&SC and EMBO Short-Term Fellowship 7232. .R. M. acknowledges

FCT for her work contract under the Transitional Rule DL 57/2016 (CTTI-57/18-I3BS(5)). J. M. O. for

his distinction attributed under the FCT Investigator program (IF/00423/2012 and IF/01285/2015);

D.C. and S.C.K. for the Portuguese Foundation for Science and Technology (FCT) under the scope of

the project 2MATCH (PTDC/BTMORG/28070/2017) funded by the Programa Operacional Regional do

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

245

Norte supported by European Regional Development Funds (ERDF); partially supported by the IET

Harvey Engineering Research Award 2018 (ENG ThE CANCER) and the European Union Framework

Program for Research and Innovation Horizon 2020 on FoReCaST project under grant agreement no.

668983.

VIII-7. REFERENCES

1. Siegel, R.L., K.D. Miller, and A. Jemal, Cancer statistics, 2018. 2018. 68(1): p. 7-30. 2. Wu, H., et al., Hydroxyethyl starch stabilized polydopamine nanoparticles for cancer chemotherapy. Chemical Engineering Journal, 2018. 349: p. 129-145. 3. Senapati, S., et al., Controlled drug delivery vehicles for cancer treatment and their performance. Signal Transduction and Targeted Therapy, 2018. 3(1): p. 7. 4. Sztandera, K., M. Gorzkiewicz, and B. Klajnert-Maculewicz, Gold Nanoparticles in Cancer Treatment. Mol Pharm, 2018. 5. Alexis, F., et al., Factors affecting the clearance and biodistribution of polymeric nanoparticles. Mol Pharm, 2008. 5(4): p. 505-15. 6. Bae, K.H., H.J. Chung, and T.G. Park, Nanomaterials for cancer therapy and imaging. Molecules and cells, 2011. 31(4): p. 295-302. 7. Kedar, U., et al., Advances in polymeric micelles for drug delivery and tumor targeting. Nanomedicine, 2010. 6(6): p. 714-29. 8. Palmerston Mendes, L., J. Pan, and V.P. Torchilin, Dendrimers as Nanocarriers for Nucleic Acid and Drug Delivery in Cancer Therapy. Molecules (Basel, Switzerland), 2017. 22(9): p. 1401. 9. Yu, B., et al., Receptor-targeted nanocarriers for therapeutic delivery to cancer. Molecular membrane biology, 2010. 27(7): p. 286-298. 10. Kopansky, E., Y. Shamay, and A. David, Peptide-directed HPMA copolymer-doxorubicin conjugates as targeted therapeutics for colorectal cancer. J Drug Target, 2011. 19(10): p. 933-43. 11. Sharma, M., et al., Folic acid conjugated guar gum nanoparticles for targeting methotrexate to colon cancer. J Biomed Nanotechnol, 2013. 9(1): p. 96-106. 12. Sarfati, G., et al., Targeting of polymeric nanoparticles to lung metastases by surface-attachment of YIGSR peptide from laminin. Biomaterials, 2011. 32(1): p. 152-161. 13. Kumazoe, M., et al., 67-kDa laminin receptor increases cGMP to induce cancer-selective apoptosis. J Clin Invest, 2013. 123(2): p. 787-99. 14. Vacca, A., et al., Melanocyte tumor progression is associated with changes in angiogenesis and expression of the 67-kilodalton laminin receptor. Cancer, 1993. 72(2): p. 455-61. 15. Shi, Y.E., et al., Expression of 67 kDa laminin receptor in human breast cancer cells: regulation by progestins. Clin Exp Metastasis, 1993. 11(3): p. 251-61. 16. Carvalho, M.R., et al., A semiautomated microfluidic platform for real-time investigation of nanoparticles' cellular uptake and cancer cells' tracking. Nanomedicine (Lond), 2017. 12(6): p. 581-596. 17. Cerqueira, S.R., et al., Multifunctionalized CMCht/PAMAM dendrimer nanoparticles modulate the cellular uptake by astrocytes and oligodendrocytes in primary cultures of glial cells. Macromol Biosci, 2012. 12(5): p. 591-7.

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

246

18. Valodkar, M., et al., Biocompatible synthesis of peptide capped copper nanoparticles and their biological effect on tumor cells. Materials Chemistry and Physics, 2011. 128(1): p. 83-89. 19. Oliveira, J.M., et al., Surface Engineered Carboxymethylchitosan/Poly(amidoamine) Dendrimer Nanoparticles for Intracellular Targeting. Advanced Functional Materials, 2008. 18(12): p. 1840-1853. 20. Maia, F.R., et al., Effect of Cell Density on Mesenchymal Stem Cells Aggregation in RGD-Alginate 3D Matrices under Osteoinductive Conditions. Macromolecular Bioscience, 2014. 14(6): p. 759-771. 21. Rea, V.E., et al., 67 kDa laminin receptor: structure, function and role in cancer and infection. Infez Med, 2012. 20 Suppl 2: p. 8-12. 22. Jiménez-Fonseca, P., et al., Gemcitabine plus capecitabine (Gem–Cape) biweekly in chemorefractory metastatic colorectal cancer. Clinical & Translational Oncology, 2015. 17(5): p. 384-392. 23. Carvalho, M.R., et al., Colorectal tumor-on-a-chip system: A 3D tool for precision onco-nanomedicine. Sci Adv, 2019. 5(5): p. eaaw1317. 24. Rudolf, E. and M. Cervinka, Membrane blebbing in cancer cells treated with various apoptotic inducers. Acta Medica (Hradec Kralove), 2005. 48(1): p. 29-34. 25. Ardini, E., et al., Identification of a novel function for 67-kDa laminin receptor: increase in laminin degradation rate and release of motility fragments. Cancer Res, 2002. 62(5): p. 1321-5. 26. Cisterna, B.A., et al., Targeted nanoparticles for colorectal cancer. Nanomedicine (Lond), 2016. 11(18): p. 2443-56.

VIII-8. SUPPLEMENTARY FIGURE

Supplementary Figure VIII-1 - Nanoparticles characterization. (A) AFM images of the CMCht/PAMAM and YIGSR-

CMCht/PAMAM NPs with respective 3D projection; (B) Peptide determination by UV spectroscopy; (C) Flow

Cytometry graphic showing selective binding of CD44 antibody to HCT-116 in a co-culture suspension with

L929 cells; (D) SEM images of a co-culture of HCT-116 cells and L929 cells: close ups showing the

nanoparticles do not agglomerate on the cell’s surface, but are widely internalized instead; and (E) Confocal

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

247

image of HCT-116 (red) and L929 cells (blue) in 3D, in the presence of FITC-YIGSR-CMCht/PAMAM NPs at

24 hours.

Supplementary Figure VIII-2 - H1NMR analysis of YIGSR peptide, CMCht/PAMAM, and YIGSR-CMCht/PAMAM. A) H1NMR

(D2O, 400 MHz) spectra of the peptide YIGSR with its chemical structure and main peaks. B) H 1NMR (D2O,

400 MHz) spectra of CMCht/PAMAM with peak assignments. C) H1NMR (D2O, 400 MHz) spectra of the

modified YIGSR-CMCht/PAMAM and respective peak assignment. D) H1NMR (D2O, 400 MHz) spectra of: the

modified YIGSR-CMCht/PAMAM (D1); CMCht/PAMAM dendrimer (D2), and theYIGSR (D3), with arrows

indicating the main peaks with the peaks marked the color according to the respective provenance.

Supplementary Figure VIII-3 - Immunocytochemistry of anti-67kDa LR (green). Pannel A shows 488 filter (green) for the 67

kDa receptor, and panel B shows 594 filter (red) performed on a co-culture of HCT-116 CRC cells and L929

Chapter VIII – Peptide-modified Dendrimer Nanoparticles for Targeted Therapy of Colorectal Cancer

248

fibroblasts ( previsouly stained in red) cells at 72 hours, proving that the cluster of HCT-116 overexpressing

67 kDa receptor consists in fact of cancer cells, and not L929.

Chapter IX –Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

249

Chapter IX

Patent – Enzymatically

Crosslinked Silk Fibroin hydrogel

microfluidic platform, methods of

production and uses thereof

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

250

Chapter IX

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin

Hydrogel Microfluidic platform, Methods of Production and Uses

Thereof

ABSTRACT

The present disclosure relates to a transparent enzymatically crosslinked silk fibroin hydrogel-based

microfluidic device that may be used ex vivo and in vivo, in tissue engineering applications, organ or

tissue disease models, tissue-, organ- and body-on-a-chip, drug discovery, drug screening, tissue

implant, tissue regeneration, and implantable microdevices.

1This chapter is based on the following publication:

M. R. Carvalho, D. Caballero, C. R. Carvalho, J. B. Costa, V. Ribeiro, Oliveira J. M., Kundu S. C., Reis R. L. Enzymatically Crosslinked Silk

Fibroin hydrogel microfluidic platform, methods of production and uses thereof. (Filed, 2019).

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

251

IX-1. TECHNICAL FIELD

The present disclosure relates to an implantable enzymatically crosslinked silk fibroin amorphous

hydrogel-based microfluidic device and methods of producing said device.

The disclosed transparent enzymatically crosslinked silk fibroin hydrogel-based microfluidic device

may be used, ex vivo and in vivo, in tissue engineering applications, organ or tissue disease models,

tissue-, organ- and body-on-a-chip, drug discovery, drug screening, tissue implant, tissue regeneration,

and implantable microdevices.

IX-2. BACKGROUND

Biomedical research has increasingly moved towards the design, fabrication and implementation of

microfluidic-based systems to efficiently improve point-of-care technologies such as drug discovery or

drug screening, diagnostics, tissue engineering and regenerative medicine. The advent of microfluidic

field has enabled precise fluidic manipulation and control of extremely small volumes. Due to its unique

properties, poly(dimethylsiloxane) (PDMS) have typically been employed in the development of

microfluidic systems. PDMS revolutionized traditional microfluidics which was previously based on

silicon and glass. However, more biologically relevant, synthetic and natural hydrogel materials have

recently been utilized in microfluidics fabrication to address the limitations of PDMS. PDMS has limited

mechanical and biochemical properties. Typical hydrogels include collagen, gelatin, gelatin

methacrylate (GelMA), agarose, alginate and other extracellular matrix proteins [1-4]. These hydrogel

materials are cell-compatible, enabling better interfaces, namely with three-dimensional cell cultures.

Nevertheless, the mechanical properties of the existing hydrogel-based microfluidic materials such

as their stiffness, only covers a limited range of native tissues, which limits their utility in mimicking the

physiological conditions of the native tissue. Additionally, the degradability of the reported microfluidic

hydrogel materials is far from desired. This is mainly because they are either not degradable or are only

stable over a limited range of time frames thus challenging their use in tissue regeneration.

Microfluidic fabrication typically involves non-degradable materials including silicon and

polydimethylsiloxane (PDMS) [5]. However, these devices are generally not implantable mainly because

they are non-degradable, and can induce inflammation and trigger foreign body reaction. Strategies for

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

252

developing more natural implantable systems comprise the use of agarose, gelatin, collagen and

alginate [6-10]. However, they do not present the best tunable stiffness properties. They are not flexible

enough to mimic the desired tissue. More recently, new biomaterials such as silk fibroin have been

employed for tissue engineering applications. Recently, silk fibroin has been used for the development

of primitive microfluidic systems. The combination of microfluidics and silk fibroin hydrogel offers a lot

of advantages. Silk fibroin is FDA approved and has gained a lot of attention in the tissue engineering

and regenerative medicine due to its excellent mechanical, biochemical, and cellular properties coupled

with its biocompatibility, flexibility, degradation properties, water-based processing, and the presence of

easily accessible chemical groups for functional modifications. This makes silk fibroin an ideal

biomaterial for the development of implantable microfluidic systems [11, 12].

A novel class of enzymatically crosslinked silk fibroin hydrogels have recently been reported but

their exploitation has mostly been limited to engineering in vitro 3D extracellular matrix (ECM) mimics

for cancer research studies (e.g. cancer invasion, migration, proliferation, and others), bioinks, and as

conduits for peripheral nerve regeneration applications [12, 13]. The horseradish peroxidase

(HRP)/hydrogen peroxide (H2O2) cross-linking approach is used in polymers containing, or

functionalized, with phenol group-containing molecules, including tyrosine, tyramine or aminophenol

[14].

Very few works describing the use of silk fibroin as an implantable material have been reported in

literature. Bettinger et al. [14] reported the fabrication of primitive microfluidic devices made by

laminating water-stable micro-moulded silk fibroin membranes in β-sheet which were modified with

macroscopic fluidic connections [15].

Document WO 2008/108838 A3 describes silk-based systems and devices such as microfluidic

devices and methods for fabricating the same. However, in this document, silk solution is simply

allowed to solidify on the molds. No enzymatic crosslinking is employed to form a hydrogel. The

resulting device is therefore in a film or layer of silk, resulting in β-Sheet conformation after initial

processing.

Document US008975073B2 describes a microfluidic device comprising silk films coupled together

to form a microchannel. However, it is described that the devices are fabricated by laminating micro-

molded and flat silk fibroin layers. Microfluidic layers are stacked, aligned, and bonded together at 70

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

253

°C, for 18 hours, under mechanical pressure, Additional 8 % aqueous silk fibroin solution is used at the

interface of the microfluidic layers. After the assembly was completed, the device was incubated in a 37

°C oven for 5 to 10 minutes to completely melt the gelatin, which was subsequently removed by

flushing the channel with deionized water. The method described in this document is substantially

different from the method disclosed in the present disclosure as the method disclosed in this document

does not involve the use of silk fibroin solution at different concentrations and different ratios.

Document WO 2010/123945 A2 document describes silk fibroin hydrogels and uses thereof. This

document discloses purified silk fibroin and method of purifying silk fibroins. This document also

discloses hydrogels comprising silk fibroin with or without an amphiphilic peptide and methods for

making hydrogels comprising silk fibroin.

Document WO18025186 describes enzymatic cross-linking of silk fibroin using horse radish and

hydroxide peroxidase to obtain a hydrogel. However, the hydrogel obtained is only an intermediate step

in the process of fabricating the final crystalline and tubular silk conduits disclosed. The silk fibroin

hydrogel described in this document is therefore an intermediate product of nerve guidance conduits.

These facts are disclosed in order to illustrate the technical problem addressed by the present

disclosure.

IX-3. GENERAL DESCRIPTION

The present disclosure relates to an implantable enzymatically crosslinked silk fibroin hydrogel-

based microfluidic device and methods of producing said device.

The disclosed enzymatically crosslinked silk fibroin hydrogel-based microfluidic device may be used

in tissue engineering applications, organ or tissue disease models, drug discovery, drug screening,

tissue implant, tissue regeneration or as implantable microdevices.

The present disclosure relates to a new methodology that comprises the use of an enzymatically

cross-linked silk fibroin hydrogel for the microfabrication of a flexible, elastic and biodegradable 3D

microfluidic chip with implantable characteristics. This methodology overcomes the problems of

previously developed PDMS microfluidic platforms.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

254

Specifically, this methodology overcomes the issues previously described by enzymatically cross-

linking silk fibroin hydrogel, using horseradish peroxidase as enzyme and H2O2 as enzyme substrate to

modify the silk fibroin solution into a hydrogel.

Silk is a naturally derived protein biomaterial with excellent biocompatibility and controllable

degradation rates, thus suitable for tissue engineering and regenerative medicine applications. The new

formulation is based on rapidly responsive silk fibroin hydrogels formed by a horseradish peroxidase

(HRP) crosslinking reaction at physiological conditions, with potential use as an artificial biomimetic

three-dimensional (3D) matrix. An advantage of using the method disclosed in the present disclosure is

the ability to directly produce silk fibroin hydrogel microfluidic device in an amorphous state this

presents an opportunity to induce β-sheet conformation later, in many different ways, if necessary.

This is the first time that silk fibroin hydrogel is employed in microfluidic technology for the

production of enzymatically cross-linked silk fibroin hydrogel-based microfluidic device.

This horseradish peroxidase – H2O2 cross-linking approach allows the development of a

substantially silk fibroin-based microfluidic device made of a flexible, implantable, biocompatible and

biodegradable biomaterial. This hydrogel retains the amorphous protein structure for at least 7 days,

allowing for cell encapsulation. Silk fibroin hydrogel resolution allows for the fabrication of micro-and

nano-sized features, such as microchannels.

In an embodiment, the method of producing silk fibroin hydrogel-based microfluidic device

comprises:

(i) the use of UV-photolithography for the fabrication of microfluidic channels;

(ii) replica molding of the fabricated structures using a polymeric material (PDMS, and others such

as polycarbonate (PC), polystyrene (PS), polyvinyl chloride (PVC), polyimide (PI);

(iii) TCS-silanization and second replica molding of the microfluidic structures using a polymeric

material (PDMS, and others);

(iv) replication of the microfluidic structures using silk fibroin hydrogel;

(v) characterization of this process with respect to microstructural fidelity and cell viability.

In an embodiment, an initially produced aqueous silk solution is transformed into an amorphous

and transparent hydrogel through a peroxidase-mediated cross-link reaction.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

255

In an embodiment, silk fibroin was combined with horseradish peroxidase solution (HRP type VI,

0.84 mg/mL) and hydrogen peroxide solution (H2O2, 0.36 wt.%).

IX-4. BRIEF DESCRIPTION OF THE DRAWINGS

The following figures provide preferred embodiments for illustrating the disclosure and should not

be seen as limiting the scope of invention.

Figure IX-1 illustrates the process of producing the PDMS mold used to produce the enzymatically

crosslinked silk fibroin hydrogel-based microfluidic device.

Figure IX-2 shows ATR-FTIR spectra for the enzymatically crosslinked silk fibroin hydrogel (with 12

% concentration of silk fibroin) retaining the amorphous protein structure for at least 7 days.

Figure IX-3 demonstrates the flexibility and elasticity of the transparent enzymatically crosslinked

silk fibroin hydrogel at day 1 and at day 3.

Figure IX-4 schematically illustrates an exemplification of the structure of the enzymatically

crosslinked silk fibroin microfluidic device. In this case, including a meandering serpentine channel,

inlets and outlets. Figure IX-4A is without dimensions while Figure IX-4B is dimensions in µm.

Figure IX-5 shows SEM images of the 3D enzymatically crosslinked silk fibroin hydrogel structures

after drying using point drying. Scale bars: 500 µm (a and b), and 100 µm (c).

Figure IX-6 schematically illustrates an example of what the enzymatically crosslinked silk fibroin

hydrogel can be used for.

Figure IX-7 shows confocal microscopy images as described in the schematics of Figure IX-6:

endothelial cells seeded on the microchannel while colorectal cells are encapsulated throughout the

enzymatically crosslinked silk fibroin microfluidic device.

Figure IX-8 shows the viability of cells encapsulated in the enzymatically crosslinked silk fibroin

hydrogel-based microfluidic device. Live/dead assay was performed and observed under confocal

microscope.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

256

Figure IX-9 shows the perfusion of blue ink through the inlet (A); formation of soluble food

colouring ink gradient (B); and diffusion of ink visible in the microchannels, allowing for the formation of

diffusion gradients of drugs/nanoparticles (C).

Figure IX-10 shows liquid perfusion thought the enzymatically crosslinked silk fibroin hydrogel-

based microfluidic device’s microchannel. A) Shows perfusion of blue ink through the inlet. B1) and B2)

show magnified images of liquid perfusion inside the serpentine microchannel

IX-5. DETAILED DESCRIPTION

The present disclosure relates to an implantable enzymatically crosslinked silk fibroin microfluidic

device and methods of producing said device.

The disclosed enzymatically crosslinked silk fibroin microfluidic device may be used in tissue

engineering applications, organ or tissue disease models, drug discovery, drug screening, tissue

implant, tissue regeneration or as implantable microdevices.

In an embodiment, the method for producing a flexible, elastic, biodegradable and implantable

microfluidic device made substantially of enzymatically crosslinked silk fibroin hydrogel comprises the

following:

Produce a SU-8 master mold with desired microchannels.

Add polymethylsiloxane (PDMS) solution at a ratio of 10:1 pre-polymer:crosslinker (vpre-polymer/vcrosslinker,

and/or wpre-polymer/wcrosslinker) to the SU-8 master mold and allow the PDMS to cure in order to obtain a PDMS

mold which is a negative master replica of the SU-8 master mold.

Silanization of the PDMS negative master mold. This process is intended to passivate the PDMS

surface to allow the fabrication of a second (positive) PDMS replica.

Pour PDMS solution at a ratio of 10:1 pre-polymer:crosslinker (vpre-polymer/vcrosslinker, and/or wpre-

polymer/wcrosslinker) on top of the TCS-silanized negative PDMS master, degas, and cure until it is completely

cured.

Peel off the negative PDMS mold to obtain a positive PDMS master mold.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

257

Prepare an aqueous silk fibroin solution with a concentration of at least 3 % (wt%), preferably from

5-25 % (wt%), more preferably 15-16 %.

Add horseradish peroxidase and hydrogen peroxide to the aqueous silk fibroin solution to form an

enzymatically cross-linked silk fibroin hydrogel;

Add silk fibroin solution at the desired concentration into the positive PDMS mold and incubate to

form a hydrogel membrane.

In an embodiment, the SU-8 master mold is produced using UV-photolithography.

In an embodiment, the SU-8 master mold is preferably produced on a silicon wafer.

In an embodiment, the PDMS is added to the SU-8 master mold and cured for at least 3 hours,

preferably for a duration from 3 hours to 12 hours, at about 37 °C.

In an embodiment, the salinization of the PDMS mold is by vapour phase method or aqueous

method, preferably by trichloro(1H,1H,2H,2H-perfluorooctyl)silane (TCS) method. The purpose of the

TCS layer is to aid the subsequent removal of the negative PDMS mold by preventing it from adhering to

the first SU-8 master mold. In particular: place the PDMS mold in vacuum (e.g., desiccator) in the

presence of a drop (drop volume from 5 µL to 100 µL) of pure TCS silanizing agent at room

temperature. Afterwards, release from vacuum and store at about 70 °C for at least 1 hour.

In an embodiment, the PDMS solution is allowed to cure on the TCS-silanized negative PDMS

master for at least 3 hours, preferably for a duration from 3 hours to 12 hours, at about 37 °C, until

the solution is completely cured.

In an embodiment, the silk fibroin solution is added to the positive PDMS mold and incubated for

at least 1 hour, preferably for a duration from 1 hour to 5 hours, at about 37 °C.

Figure IX-1 illustrates the process of producing the PDMS mold used to produce the silk fibroin

hydrogel-based microfluidic device.

In an embodiment for better results, the thickness of the enzymatically crosslinked silk fibroin

hydrogel layer can be controlled by controlling the volume of silk solution poured into the PDMS molds.

In an embodiment, for better results, the mechanical properties of the silk fibroin hydrogel can be

tuned by changing the concentration of silk fibroin proteins. In an embodiment, for better results, the

silk fibroin hydrogel is enzymatically cross-linked with horseradish peroxidase and hydrogen peroxide,

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

258

but other peroxidases and oxidizers may be used. In an embodiment, for better results, the silk hydrogel

is functionalized with drugs or chemo-attractants.

In an embodiment, for better results, the hydrogel comprises from 3 % up to 25 % (wt %) of silk

fibroin, preferably from 8% to 20 % (wt %) of silk fibroin, more preferably from 10% to 12 % (wt %) of silk

fibroin.

In an embodiment, an initially produced aqueous silk solution is transformed into an amorphous

and transparent hydrogel, through a peroxidase-mediated cross-linking reaction.

In an embodiment, the enzymatically crosslinked silk fibroin hydrogel retains the amorphous

protein structure for at least 7 days, allowing for cell encapsulation.

Figure IX-2 shows ATR-FTIR spectra for the 3D enzymatically crosslinked silk fibroin hydrogel (with

12 % concentration of silk fibroin) retaining the amorphous protein structure for at least 7 days.

In an embodiment, for better results, the enzymatically crosslinked silk fibroin microfluidic device

further comprises a biological active agent, a therapeutic agent, an additive, a pharmaceutically

acceptable excipient, a pharmaceutically acceptable carrier, and mixtures thereof.

In an embodiment, the enzymatically crosslinked silk fibroin hydrogel-based microfluidic device

functions to mimic the extracellular matrices (ECM) of the body. For example, the silk fibroin hydrogel-

based microfluidic device may serve as a physical support and/or an adhesive substrate for isolated

cells during in vitro culture and subsequent implantation. As the transplanted cell populations grow and

the cells function normally, the cells will begin to secrete their own ECM support and the silk-based

microdevice may then biodegrade. The biodegradation of the silk fibroin hydrogel-based microfluidic

device may be controlled by various manufacturing techniques known in the art.

In an embodiment, the enzymatically crosslinked silk fibroin hydrogel-based microfluidic devices

are mechanically robust, transparent, flexible, elastic and possess microchannels with different

geometries and sizes for seeding cells and flowing fluids (including body fluids) through the device.

Figure IX-3 demonstrates the flexibility and elasticity of the transparent silk fibroin hydrogel at day 1

and at day 3.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

259

Figure IX-4 schematically illustrates an exemplification of the structure of the enzymatically

crosslinked silk fibroin microfluidic device. In this case, including a serpentine channel, inlets and

outlets, without (A) and with (B) dimensions in µm.

Figure IX-5 Scanning Electron Microscope (SEM) images of the 3D silk hydrogel structures after

drying using critical point drying. Scale bars: 500 µm (a and b), and 100 µm (c).

In an embodiment, the enzymatically crosslinked silk fibroin hydrogel-based microfluidic devices

are fabricated to support the growth of cells, including but not limited to eukaryotic cells (cancer cells,

stromal cells, immune system, or endothelial cells). Other cell types could also be added as required.

In an embodiment, a portion of a surface of the microchannels in the silk fibroin hydrogel-based

microfluidic device supports cell growth, in their native morphology.

In an embodiment, a portion of the enzymatically crosslinked silk fibroin hydrogel-based

microfluidic device supports cell growth by encapsulating cells.

Figure IX-6 schematically illustrates an example of what can be done using the hydrogel. For

example, colorectal cancer microenvironment can be mimicked by seeding human colonic

microvascular endothelial cells inside the microchannels, invading the matrix in response to VEGF

gradients, while HCT-116 colorectal cells are encapsulated in the silk.

In an embodiment, the enzymatically crosslinked silk fibroin hydrogel-based microfluidic devices

are fabricated to support the growth of cells, including but not limited to eukaryotic cells (cancer cells,

stromal cells, immune system, or endothelial cells). Other cell types could also be added as required.

Figure IX-7 shows the confocal microscopy images as described in the schematics of Figure IX-6:

endothelial cells seeded on the microchannel, while colorectal cells are encapsulated throughout the

enzymatically crosslinked silk hydrogel microfluidic device.

Figure IX-8 shows the viability of cells encapsulated in the enzymatically crosslinked silk fibroin

hydrogel-based microfluidic device. Live/dead assay was performed and observed under confocal

microscope. Live/dead assay was performed and observed under confocal microscope. Live cells are

stained in green, while dead cells are stained in red.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

260

Figure IX-9 shows the perfusion of blue ink through the inlet (A); formation of soluble ink gradient

(B); and diffusion of food colouring ink visible in the microchannels, allowing for the formation of

diffusion gradients of drugs/nanoparticles (C).

In an embodiment, oscillatory rheological measurements were performed to determine the

rheological properties of the enzymatically cross-linked silk fibrin hydrogels at different concentration (6

%, 12 %, and 14 % of silk solution). The results were compared to PDMS.

Figure IX-10 shows liquid perfusion through the enzymatically crosslinked silk fibroin hydrogel-

based microfluidic device’s microchannel. (A) Shows perfusion of blue ink through the inlet. (B1) and

(B2) show magnified images of liquid perfusion inside the serpentine microchannel.

Table IX-1 - Composition, rheological properties of hydrogels at a frequency of 0.1 Hz.

Name Cells Incubation

Time (minutes)

Storage modulus

(G’, Pa)

Loss modulus

(G'', Pa)

6 % no 30 1963±151 30±16

12 % no 30 6585 ± 253 152 ± 48

14 % no 30 7173 ± 605 387 ± 64

The term "comprising" whenever used in this document is intended to indicate the presence of

stated features, integers, steps, components, but not to preclude the presence or addition of one or

more other features, integers, steps, components or groups thereof.

The disclosure should not be seen in any way restricted to the embodiments described and a

person with ordinary skill in the art will foresee many possibilities to modifications thereof.

The above described embodiments are combinable.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

261

IX-6. CLAIMS

1. An enzymatically cross-linked silk fibroin hydrogel microfluidic device; wherein said device is

flexible and elastic; and wherein said microfluidic device comprises a microchannel configured to allow

liquid media to flow.

2. The enzymatically cross-linked silk fibroin microfluidic device according to the previous claim

wherein the hydrogel is transparent, preferably for at least 7 days.

3. The enzymatically cross-linked silk fibroin microfluidic device according to any of the previous

claims wherein the hydrogel is biodegradable.

4. The enzymatically cross-linked silk fibroin microfluidic device according to any of the previous

claims wherein the device is implantable.

5. The enzymatically cross-linked silk fibroin microfluidic device according to any of the previous

claims wherein the silk fibroin hydrogel retains its amorphous protein structure for at least 7 days.

6. The enzymatically cross-linked silk fibroin microfluidic device according to any of the previous

claims wherein the enzymatic cross-linking of the silk fibroin hydrogel is by an enzyme horseradish

peroxidase and an oxidizer hydrogen peroxide.

7. The enzymatically cross-linked silk fibroin microfluidic device according to any of the previous

claims wherein the concentration of silk fibroin is from 3 % to 25 % (wt.%) of silk fibroin, preferably from

8 % to 20 % (wt %) of silk fibroin, more preferably from 10 % to 12 % (wt %) of silk fibroin.

8. The enzymatically cross-linked silk fibroin microfluidic device according to any of the previous

claims wherein the device further comprises extracellular matrix, growth factor, drug, cell, and

combinations thereof.

9. The enzymatically cross-linked silk fibroin microfluidic device according to any of the previous

claim wherein the cell is selected from the list: cancer cell, stromal cell, immune system, endothelial

cell, or combinations thereof.

10. The enzymatically cross-linked silk fibroin microfluidic device according to any of the previous

claims further comprising:

an inlet for liquid media to enter the microfluidic device;

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

262

an intermediate meandering section for liquid media to flow;

a first microchannel connecting the inlet and the intermediate meandering microchannel;

an outlet for liquid media to exit the microfluidic device;

a second microchannel connecting the outlet and the intermediate meandering part.

11. The enzymatically cross-linked silk fibroin microfluidic device according to the previous claim

wherein the length of the device is from 150 mm to 5 mm, preferably 75 mm to 10 mm, more

preferably 42 mm.

12. The enzymatically cross-linked silk fibroin microfluidic device according to the previous claim

wherein the width of the inlet and the outlet are each from 10 mm to 200 μm, preferably 5 mm to 500

μm, more preferably 1 mm.

13. The enzymatically cross-linked silk fibroin microfluidic device according to the previous claim

wherein the thickness of the intermediate meandering section from 20 mm to 1 mm, preferably 10 mm

to 2 mm, more preferably 6.5 mm.

14. The enzymatically cross-linked silk fibroin microfluidic device according to the previous claim

wherein the width of the microchannel is from 10 μm to 2 mm, preferably 100 μm to 1 mm, more

preferably 200 μm.

15. An enzymatically cross-linked silk fibroin hydrogel for use in medicine or veterinary, wherein

the enzymatically cross-linked silk fibroin hydrogel is administrated in the form of an implantable

microfluidic device.

16. The enzymatically cross-linked silk fibroin hydrogel according to the previous claim for use in

tissue engineering, organ disease models, tissue disease models, drug discovery, drug screening, tissue

implant or tissue regeneration.

17. A Kit comprising an implantable microfluidic device or a hydrogel according to any of the

previous claims.

18. The kit or the hydrogel according to the previous claim wherein the kit further comprises at

least one of the following components: Cells, growth factor, therapeutic agent, and mixtures thereof.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

263

19. A method of producing the implantable microfluidic device according to any of the previous

claims comprising the following steps:

obtaining a polymethylsiloxane silanized master mold with predetermined microchannels;

preparing an enzymatically cross-linked silk fibroin hydrogel with horseradish peroxidase and

hydrogen peroxide;

adding the silk fibroin hydrogel into the positive PDMS mold and incubate to form a hydrogel

membrane.

20. The method of producing the implantable microfluidic device according to the previous claim

comprising the following steps:

obtaining a master mold with predetermined microchannels, preferably a SU-8 master mold;

adding a polymethylsiloxane (PDMS) solution, preferably at a ratio of 10:1 pre-polymer:crosslinker,

(vpre-polymer/vcrosslinker, and/or wpre-polymer/wcrosslinker), to the master mold and allow the PDMS to cure in order to obtain

a PDMS negative master mold which is a negative master mold replica; silanizing of the PDMS negative

master mold to obtain a TCS-silanized negative PDMS master; pouring a second PDMS solution,

preferably at a ratio of 10:1 pre-polymer:crosslinker (vpre-polymer/vcrosslinker, and/or wpre-polymer/wcrosslinker) on top of the

TCS-silanized negative PDMS master, degas, and cure;

peeling off the negative PDMS mold to obtain a positive PDMS master mold;

preparing an enzymatically cross-linked silk fibroin hydrogel with horseradish peroxidase and

hydrogen peroxide;

adding the silk fibroin hydrogel into the positive PDMS mold and incubate to form a hydrogel

membrane.

21. The method of producing the implantable microfluidic device according to any of the previous

claims 21-22 wherein the aqueous silk fibroin solution has a silk fibroin concentration of at least 3 %

(wt.%), preferably from 3-25 % (wt.%), more preferably from 8 to 20 % (wt.%), more preferably from 10 to

12 % (wt.%).

22. The method of producing the implantable enzymatically cross-linked silk fibroin microfluidic

device according to any of the previous claims 21-23 wherein the SU-8 master mold is produced using

UV-photolithography.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

264

23. The method of producing the enzymatically cross-linked silk fibroin implantable microfluidic

device according to any of the previous claims 21-24 wherein the PDMS is added to the SU-8 master

mold and cured for at least 3 hours, preferably for a duration from 3 hours to 12 hours, at about 37

°C.

24. The method of producing the implantable enzymatically cross-linked silk fibroin microfluidic

device according to any of the previous claims 21-25 wherein the PDMS solution is allowed to cure on

the TCS-silanized negative PDMS master for at least 3 hours, preferably for a duration from 3 hours to

12 hours, at about 37 °C, until the solution is completely cured.

25. The method of producing the implantable enzymatically cross-linked silk fibroin microfluidic

device according to any of the previous claims 21-26 wherein the silk fibroin solution is added to the

positive PDMS mold and incubated for at least 1 hour, preferably for a duration from 1 hour to 5 hours,

at about 37 °C.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

265

IX-7. LIST OF FIGURES

Figure IX-1 - Illustration the process of producing the PDMS mold used to produce the enzymatically crosslinked silk fibroin

hydrogel-based microfluidic device.

Figure IX-2 – ATR-FTIR spectra for the enzymatically crosslinked silk fibroin hydrogel (with 12 % concentration of silk fibroin)

retaining the amorphous protein structure for at least 7 days.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

266

Figure IX-3 – Demonstration of the flexibility and elasticity of the transparent enzymatically crosslinked silk fibroin hydrogel at

day 1 and at day 3.

Figure IX-4 - Schematically illustrates an exemplification of the structure of the enzymatically crosslinked silk fibroin

microfluidic device. In this case, including a meandering serpentine channel, inlets and outlets. Figure 4A is

without dimensions while Figure 4B is dimensions in µm.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

267

Figure IX-5 - SEM images of the 3D enzymatically crosslinked silk fibroin hydrogel structures after drying using point drying.

Scale bars: 500 µm (a and b), and 100 µm (c).

Figure IX-6 - Schematically illustration of an example of what the enzymatically crosslinked silk fibroin hydrogel can be used

for.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

268

Figure IX-7 - Confocal microscopy images as described in the schematics of figure 6: endothelial cells seeded on the

microchannel while colorectal cells are encapsulated throughout the enzymatically crosslinked silk fibroin

microfluidic device.

Figure IX-8 - Viability of cells encapsulated in the enzymatically crosslinked silk fibroin hydrogel-based microfluidic device.

Live/dead assay was performed and observed under confocal microscope.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

269

Figure IX-9 - Perfusion of blue ink through the inlet (A); formation of soluble food colouring ink gradient (B); and diffusion of

ink visible in the microchannels, allowing for the formation of diffusion gradients of drugs/nanoparticles.

Figure IX-10 - Liquid perfusion through the enzymatically crosslinked silk fibroin hydrogel-based microfluidic device’s

microchannel. (A) Shows perfusion of blue ink through the inlet. (B1) and (B2) show magnified images of

liquid perfusion inside the serpentine microchannel (or meandering section).

IX-8. REFERENCES

1. Buchanan, C.F., et al., Three-dimensional microfluidic collagen hydrogels for investigating flow-mediated tumor-endothelial signaling and vascular organization. Tissue engineering. Part C, Methods, 2014. 20(1): p. 64-75.

2. Liu, J., et al., Hydrogels for Engineering of Perfusable Vascular Networks. International journal of molecular sciences, 2015. 16(7): p. 15997-16016.

3. Johann, R. and P. Renaud, Microfluidic patterning of alginate hydrogels. Vol. 2. 2007. 73-9. 4. Lee, Y., et al., Photo-crosslinkable hydrogel-based 3D microfluidic culture device: Microfluidics and Miniaturization. Vol.

36. 2015. 5. Chen, C., et al., 3D-printed Microfluidic Devices: Fabrication, Advantages and Limitations-a Mini Review. Anal Methods,

2016. 8(31): p. 6005-6012. 6. Chung, B.G., et al., Microfluidic fabrication of microengineered hydrogels and their application in tissue engineering.

Lab Chip, 2012. 12(1): p. 45-59.

Chapter IX - Patent - Enzymatically Crosslinked Silk Fibroin Hydrogel Microfluidic platform, Methods of Production and Uses Thereof

270

7. Ling, Y., et al., A cell-laden microfluidic hydrogel. Lab on a Chip, 2007. 7(6): p. 756-762. 8. Bertassoni, L.E., et al., Hydrogel bioprinted microchannel networks for vascularization of tissue engineering constructs.

Lab on a Chip, 2014. 14(13): p. 2202-2211. 9. Gillette, B.M., et al., In situ collagen assembly for integrating microfabricated three-dimensional cell-seeded matrices.

Nat Mater, 2008. 7(8): p. 636-40. 10. Choi, N.W., et al., Microfluidic scaffolds for tissue engineering. Nat Mater, 2007. 6(11): p. 908-15. 11. Carvalho, M.R., et al., Microfluidics: Tuning Enzymatically Crosslinked Silk Fibroin Hydrogel Properties for the

Development of a Colorectal Cancer Extravasation 3D Model on a Chip (Global Challenges 5-6/2018). Global Challenges, 2018. 2(5-6): p. 1870164.

12. Carvalho, C.R., et al., Tunable Enzymatically Cross-Linked Silk Fibroin Tubular Conduits for Guided Tissue Regeneration. 2018. 7(17): p. e1800186.

13. Costa, J.B., et al., INKS FOR 3D PRINTING, METHODS OF PRODUCTION AND USES THEREOF. Patent WO/2018/225049.

14. Teixeira, L.S., et al., Enzyme-catalyzed crosslinkable hydrogels: emerging strategies for tissue engineering. Biomaterials, 2012. 33(5): p. 1281-90.

15. Bettinger, C.J., et al., Silk Fibroin Microfluidic Devices. Adv Mater, 2007. 19(5): p. 2847-2850.

271

SECTION 4

Conclusions and Future

Perspectives

Chapter X - Conclusions and Future Perspectives

272

Chapter X

Conclusions and Future

Perspectives

Chapter X - Conclusions and Future Perspectives

273

Chapter X

Chapter X - Conclusions and Future perspectives

X-1. GENERAL CONCLUSIONS

Colorectal cancer represents the most prevalent cause of cancer-related deaths, mainly resulting

from its metastatic nature. Conventional adjuvant therapies for colorectal cancer including chemo,

radiotherapies and biological agents have improved cancer therapy, but they carry serious side effects

such as the development of drug resistance and toxicity to healthy cells. All of the above are strong

indicators suggesting that we need better strategies to fight colorectal cancer. More importantly, science

needs to move forward with the development and screening of new chemotherapeutic drugs, since in

the particular case of cancer drugs, studies report that the average rate of successful translation from

animal models to clinical cancer trials is less than 8 %.

The major goal of the work developed under the scope of the present thesis was to develop new

strategies to improve the state of the art in what regards 3D microfluidic in vitro models for colorectal

cancer research, with a strong approach on dendrimer nanoparticles and natural-origin polymers, such

as carboxymethilchitosan and silk fibroin.

To improve the state of the art, the development of dendrimer nanoparticles (targeted and not

targeted); the understanding of the fundamental differences between cancer research experiments in

2D vs. 3D; the differences between static conditions vs. dynamic conditions, and the exploration of new

extracellular matrix-like materials to be used as alternative to Matrigel®, seems like a promising option

leading to a successful outcome. These were the goals of the first experimental chapters of this thesis.

All the technologies studied and proposed procured to match the main drawbacks identified in the

literature and attempted to overcome them, recurring to innovative and more precise technologies.

In Chapter V, the goal of this work was to develop a platform composed of labeled dendrimer

nanoparticles and a microfluidic device for real-time monitoring of cancer cells’ fate. Also, it was

important to study the influence of a dynamic environment (e.g. perfusion of FITC-labeled

carboxymethylchitosan/poly(amidoamine) dendrimer nanoparticles in the internalization levels and

potential cytotoxicity in three different cancer cell lines (glioblastoma, colorectal and cervical cancer).

Chapter X - Conclusions and Future Perspectives

274

When compared to the traditional 2D culture flasks, cells cultured under perfusion showed higher

sensitivity to the nanoparticles. Reports in the literature had shown the opposite result, being gold

nanoparticles less toxic e dynamic conditions. This is evidence that efforts have to be done in order to

improve our experimental designs in an attempt to get closer to more relevant (in vivo like) conditions.

After the extensive characterization performed on the dendrimer nanoparticles, including differential

scanning calorimetry (that had never been performed in carboxymethylchitosan/poly(amidoamine)),

further biological work led to the conclusion that a carboxymethylchitosan/poly(amidoamine) dendrimer

concentration of 0.5 mg/mL was ideal for further studies, since it was demonstrated not to be cytotoxic

in dynamic conditions. By the end of this first chapter, the proposed microfluidic-based was validated

for real-time monitoring of nanoparticles fate, which upon more studies, namely the assessment of an

anticancer drug release effect, could be used for cancer theranostics.

In Chapter VI, the thesis focuses on colorectal cancer, and there is an increase in the complexity of

the microfluidic platform used. Moreover, it was aimed at moving forward the general knowledge in

terms of understanding the specific contribution of the mechanical properties of materials/hydrogels

used as extracellular matrix like materials in 3D cancer models. Silk was chosen due to its unique

properties, both biologic and mechanical, as well as the ability to isolate the effect of the mechanical

properties due to the lack of mammalian motifs. Moreover, we were able to compared cancer cell

migration in the commercial Vena4™ microfluidic platform with the traditionally used Boyden chamber

assay, to assess any particular differences. 2 and 3 % enzymatically crosslinked silk fibroin hydrogels

were chosen based on a previous screening, and compared to Matrigel® (gold standard extracellular

matrix material). The conclusions of this work were striking: when comparing colorectal cancer cell

migration in 3D. Although Matrigel® is much softer (seen in rheological analysis) and often chosen as

the main extracellular matrix like material, 2 % silk hydrogel allowed more intense cellular migration

(despite the higher stiffness). Moreover, in the Boyden chamber assay the results were not similar,

being the migration levels comparable between 2 % silk fibroin and Matrigel®. Again, evidence was

shown that the models chosen to perform cancer studies play an important role in the results that are

obtained, as well as the materials. Controlling and understanding the effect of the various components

of these models pushes the field forward and enables investigation of interactions within the tumor

microenvironment, as well as the response to stimuli, such as chemoattractants and

chemotherapeutics.

Chapter X - Conclusions and Future Perspectives

275

For the purpose of Chapter VII, a very fruitful collaboration was established between 3Bs Research

Group and MERLN Institute (Maastricht University, Netherlands). The main goal was to develop a 3D

microfluidic platform with the following requirements: ease of use; mimic the colorectal tumor

complexity (vascular and tumor parts), possibility to create gradients; possibility of real-time imaging;

performance of immunocytochemistry and cell retrieval after the experiments for polymerase chain

reaction (PCR) analysis. This task was achieved using the design explored in Chapter VII, and by means

of successfully establishing a dynamic co-culture of colonic endothelial microvascular cells and HCT-

116 colorectal cancer cells embedded in Matrigel® in central chamber. Most of all, the advantages of

our platform as compared to others of the same type found in the literature (central chamber for 3D

matrix laterally sided by perfusable channels) are: 1) facile molecular cell analysis; 2) physiologically

inspired design of radial drug penetration into solid tumors; 3) hybrid co-culture system to show

efficiency of drug penetration through a micro-vascular network into a cancer-mimicking tissue,

comprised of a supporting relevant matrix with colorectal cancer cells; 4) real-time imaging of drug

delivery and its effect on cell viability; 5) design compatible with assessing physiologically-relevant cell

migration, thus metastatic events (not the scope of the current manuscript); and, 6) system easily

combinable with organoid seeding of the central chamber, which allows for the integration of other

models. In general, we were able to generate stable gradients of nanoparticles releasing gemcitabine,

resulting in a gradient of colorectal cancer cell death. Moreover, we observed a down-regulation for all

genes teste (MMP-1, Caspase-3 and Ki-67). This model was validated as a platform for precision drug

screening that can later be applied in personalized medicine.

Following the observations on Chapter VII regarding the deleterious side-effects of gemcitabine

release over the endothelial cells, it was of crucial importance the development of a strategy for

targeting the nanoparticles internalization (and its consequent drug release) in colorectal cancer cells,

for a more effective treatment and reduced toxicity. To achieve this goal, Chapter VIII aimed at

carboxymethylchitosan/poly(amidoamine) targeting towards colorectal cancer cells via interaction with

the laminin receptor, whose expression is known to be upregulated in some types of cancer (e.g. breast

and ovarian cancer). Overall, this study provided a proof-of-concept for the ability of surface attached

YIGSR, a laminin-derived peptide, to carboxymethylchitosan/poly(amidoamine) to target over-expressing

laminin receptor HCT-116 colorectal cancer cells, when in co-culture with L929 cells. This was

observed when co-cultures were seeded in traditional 2D well-plates or in a Matrigel® drop (3D). Not

only the preferential internalization was demonstrated, but also the effect of gemcitabine drug release

Chapter X - Conclusions and Future Perspectives

276

from these modified nanoparticles was assessed in these systems. Results show clear preferential

internalization of nanoparticles in cancer cells, as well as higher levels of cancer cell death when in the

presence of other non-cancerous cell types.

Chapter IX is focused on an international patent which discloses a new methodology to obtain the

transparent enzymatically crosslinked silk fibroin hydrogel-based microfluidic device, as well as their

possible applications in Tissue Engineering and biomedical field in general. The innovation of the

method relied on the transformation of an initially produced aqueous silk fibroin solution into an

amorphous and optically transparent hydrogel, through a horseradish peroxidase-mediated crosslink

reaction. Combining this technology with the use of microfabrication (namely UV-photolithography and

replica moulding) for the fabrication of microfluidic channels, it is possible to obtain a fully transparent

and flexible microfluidic silk platform. This methodology allowed overcoming the problems of previously

developed PDMS microfluidic platforms, making it implantable due to silk fibroin biocompatibility and

biodegradability. After the initial laboratorial work performed in the scope of the patent, the method for

fabricating an enzymatically cross-linked silk fibroin hydrogel microfluidic device; flexible and elastic;

comprising microchannels configured to allow liquid media to flow, was developed. This platform can

find use in medicine or veterinary, wherein the enzymatically cross-linked silk fibroin hydrogel is

administrated in the form of an implantable microfluidic device. Chemoattractants, therapeutic drugs or

other molecules of interest can be easily incorporated in the silk fibroin solution, even in the form of

gradients Therefore, it can be used in tissue engineering, organ disease models, tissue disease models,

drug discovery, drug screening, tissue implant or tissue regeneration. The developed enzymatically

crosslinked silk fibroin microfluidic platform is transparent, flexible and reproducible, allowing as well as

the possibility of modulating the mechanical properties according to the tissue that is being studied.

X-2. FUTURE RESEARCH DIRECTIONS

This thesis aimed at combining several fields of research for the purpose of moving forward and

increasing the current knowledge on cancer research, specifically colorectal cancer research. For this,

nanotechnology principles were combined with microfluidics and tissue engineering concepts. To date,

despite the remarkable advances and progresses in colorectal cancer studies, there is no ideal

microfluidic model that could guarantee future success in clinical trials, or ideal drug-loaded

nanoparticle that would undoubtedly treat colorectal cancer. One of the problems that has slowed down

Chapter X - Conclusions and Future Perspectives

277

the development and approval of new anticancer therapies is the lack of pre-clinical models that can be

used to identify key molecular, cellular and biophysical features of human cancer progression. This is

because most in vitro cancer models fail to faithfully recapitulate the native tissue and organ

microenvironment in which tumors form, which substantially contributes to the complex

pathophysiology of the disease. More complex in vitro cancer models have been developed, including

transwell cell cultures, spheroids and organoids grown within flexible extracellular matrix gels, which

better mimic normal and cancerous tissue development than cells maintained on conventional 2D

substrates. But these models still lack the tissue–tissue interfaces, organ-level structures, fluid flows

and mechanical cues that cells experience within living organs, and furthermore, it is difficult to collect

samples from the different tissue micro-compartments. Microfluidic devices are among the most

promising tools to overcome these hurdles and best mimic in vivo like conditions, either in normal or

disease scenarios, such as tumorigenesis or pathogenesis. Together with the potential of biomaterials

as artificial extracellular matrix, its combination with microfluidics represents the ability to more closely

mimic cells’ natural microenvironment, concerning its three-dimensional nature, continuous perfusion

with nutrients and cells’ crosstalk. Due to miniaturization and increased experimental throughput,

microfluidics have generated significant interest in the drug discovery and development domain. In its

turn, biomaterials are gradually being developed as in vitro microenvironments mimicking in vivo cell

niches. It was one of our main goals to explore the newly developed enzymatically crosslinked silk

fibroin hydrogels as a new biomaterial, isolating the effects of mechanical properties. It is believed that

chip-based 3D cell culture in cancer research will be largely dependent on the improvement of

biomaterials that can best emulate the extracellular matrix and the capacity to scale-up these complex

technologies. Moreover, microfluidic models need standardization and reproducible formats suitable for

high-throughput applications. If cell migration studies are to be used to test drug efficacy in this context,

these models have to be reliable in what regards its fabrication and incorporation of the most adequate

biomaterials to reduce arbitrary cell migration patterns.

Regarding the use of dendrimer nanoparticles, its future application to nanomedicine depends on

batch-to-batch reproducibility of the final materials’ properties. Therefore, to advance with dendrimer-

based materials into clinical trials it is crucial to develop well-defined multifunctional dendrimers with a

controlled number and location of drugs, targeting groups, imaging agents, solubilizing agents and/or

other relevant motifs. Recent developments on synthetic tools are defying the outdated perception that

dendrimers’ synthesis is complex, slow, tedious and cost-prohibitive. Indeed, the very low amount of

Chapter X - Conclusions and Future Perspectives

278

dendrimers to be used in therapeutic applications will most likely dilute their cost, making their

cost/benefit relationship promising for further product development. The next steps in the dendrimer

field will be very challenging and exciting, because it is time to meet the high expectations that were

placed on dendrimers.

From a practical point of view, there is a never ending myriad of possibilities when it comes to

complementing and deepening the developed work. The first one would be the use of primary cells from

colon, either normal or cancerous. No work is fully complete without testing the developed technologies

in primary cells. For this, the already established collaborations with local hospitals for the acquisition of

colorectal biopsies would be of crucial importance, making the performed tests adjustable to the proof-

of-concept approach for personalized medicine. An in silico model of the distribution of the

nanoparticles thought the central chamber would be a plus to complement Chapter VII. The real

possibility of developing a computer mathematical model correlating the exact fluorescence intensity

with drug release concentration present at each point would be of added value and could consist of a

continuation of this work. Regarding Chapter VIII, an in vivo study in mice to determine the effect of

drug loaded targeted nanoparticles in colorectal cancer induced mice would be of added value.

One particularly promising strategy to be pursued in this line of work would be the circulating

tumor cells approach. Using an implantable microfluidic device such as the one developed in this thesis

in Chapter IX, could be able to recruit and capture circulating tumor cells in the patient, being later

subjected to drugs screening. Owing to the less-invasive nature of this procedure, chemotherapy could

be tested, and routine monitoring of cancer progression could be achieved. Although promising,

implantability tests have to be performed in mouse models first, which will be the next immediate step.

As a conclusion of this thesis, although significant steps have been taken separately in what

regards the study of biomaterials in the tumorigenesis processes and cancer cell migration,

development of drug screening technologies, and nanoparticle targeting, studies involving all of the

aforementioned technologies still have to be performed to test all these technologies together.

Before the fight against cancer is finally won, the advances in in the mentioned fields will continue

to help researchers achieve a better understanding of cancer biology as well as develop novel

therapeutic strategies aiming personalized treatments.